Bureaucracy Times

All the news fit to be declassified.

  • Federal Register Watch: September 30, 2026

    The Treasury Department and the IRS published temporary regulations today spelling out how Trump Accounts, the new tax-advantaged children’s savings accounts, will actually be established and funded, including a provision letting the Treasury Secretary open an account automatically for an eligible child even if no parent has done so first.

    • Treasury Department / IRS — Trump Accounts temporary regulations lay out how the new accounts are established, including automatic enrollment of eligible children by the Treasury Secretary when a parent hasn’t opened one. That automatic-enrollment default is worth noting because it’s a different starting point than most federal benefit programs use, which typically require someone to apply; here the account can exist before a family has taken any action, raising practical questions about how parents will be notified and how they take control of an account opened on a child’s behalf.
    • National Highway Traffic Safety Administration — NHTSA finalized the SAFE Vehicles Rule III, recalibrating Corporate Average Fuel Economy standards for model years 2022 through 2031 and revising how vehicles are classified for compliance purposes. The agency says the changes are needed “to ensure compliance with law,” language that signals this final rule is closing out a legal vulnerability in how the standards were set rather than simply updating them on schedule, and the classification changes matter because they can shift which fuel-economy target a given vehicle model is actually held to.
    • Treasury Department, Office of Foreign Assets Control — OFAC tightened the Cuban Assets Control Regulations, eliminating the “U-turn” exception that let U.S. banks process certain Cuba-linked payments passing through the American financial system and ending authorization for group people-to-people and professional-meeting travel, carrying out a June 2025 presidential directive. The same day, OFAC also issued a brand-new Cuba Sanctions Regulations part implementing a separate May 2026 executive order, while noting that fuller definitions and general licenses are still to come; publishing the enforcement mechanism before the interpretive guidance that explains exactly what it covers is the kind of sequencing worth watching for how banks and travelers are expected to comply in the meantime.
    • Office of Management and Budget — OMB published the formal notice of a presidential rescissions package sent to Congress on September 25, proposing to cancel $810 million in already-appropriated funds across the Commerce, Education, Health and Human Services, Homeland Security, Housing and Urban Development, and Justice departments, plus international assistance programs. Filing this notice starts a 45-day clock under the Impoundment Control Act during which Congress can approve the cuts or let the proposal lapse, but the administration can hold the money back while that clock runs, a mechanism that has drawn scrutiny in past rescissions fights over whether it lets the executive branch withhold spending Congress already approved.
    • Treasury Department — Treasury issued an interim final rule setting the forms and procedures the new Stablecoin Certification Review Committee will use to decide whether a state’s stablecoin regulatory regime counts as “substantially similar” to the federal framework under the GENIUS Act, a call that determines whether state-chartered stablecoin issuers can operate under state rather than federal oversight. Issuing the procedures as an interim final rule means they take effect immediately without the advance public comment period ordinarily used to shape a new agency process, though Treasury says it will accept comments and may revise the rule later, worth watching given how much rides on the Committee’s decisions.
    • U.S. Citizenship and Immigration Services — USCIS finalized new EB-5 immigrant investor fees and folded in provisions from the EB-5 Reform and Integrity Act of 2022 that Congress passed but USCIS is only now codifying into fee regulations. The rule follows a proposed version published back in October 2025, and that roughly year-long gap is a useful reminder of how long even routine fee rulemakings take to finalize, regardless of how long the underlying statute has already been law.

    Editorial note: We screened all 118 documents published in the Federal Register on September 30, 2026, before selecting the items above. Excluded as routine: roughly two dozen agency information-collection notices submitted to OMB under the Paperwork Reduction Act; seven antidumping and countervailing-duty proceedings, sunset reviews, and court-decision notices; seven self-regulatory-organization (stock exchange) rule filings, including a batch of near-identical liquid-cooled-cabinet fee changes across four Nasdaq exchanges; sixteen NAGPRA repatriation and inventory-completion notices; a cluster of OFAC general-license republications spanning the Venezuela, Iran, and Russia sanctions programs; advisory-committee and closed-meeting notices; and three Presidential disaster-declaration amendments.

  • We Asked America.gov 15 Questions the Government Already Answered. It Did Surprisingly Well.

    The federal government launched a new artificial intelligence system Tuesday with an ambitious proposition: instead of making Americans figure out which agency, office, subagency, program, website, form and acronym possesses the thing they need, perhaps the government could figure that out for them.

    It is called America.gov, and on its first night in operation, Bureaucracy Times decided to give it a test.

    Not a test involving the capital of Nebraska or how to renew a passport. We gave it questions from our own reporting, including obscure federal contracts, nuclear waste, contaminated nickel, pension reporting requirements, used cooking oil, veterans benefits fraud, classified Pentagon clocks, voter records and one of the strangest securities disputes to pass through FINRA in recent years.

    The result was not what we expected.

    America.gov did remarkably well.

    Across 15 questions for which Bureaucracy Times had independently established an answer, established that no public answer exists, or possessed information unavailable to the general public, America.gov produced nine answers we classified as correct and six that were substantially correct but incomplete.

    We identified no fabricated answers.

    Not one.

    For an artificial intelligence system introduced by the federal government, an institution that has occasionally demonstrated an almost artisanal commitment to making simple information difficult to obtain, that deserves some credit.

    America.gov did have problems. It sometimes failed to retrieve information buried inside federal reports. Some of its statistics were years behind newer government figures. It occasionally found one government source while overlooking another containing the missing answer.

    And, in one particularly delightful encounter with the machinery of modern government, America.gov prevented us from asking about a federal contract because it apparently mistook the government’s own contract number for our personal information.

    Still, after an evening spent deliberately trying to make the new system stumble, the most consistent behavior we encountered was something increasingly valuable in the age of generative artificial intelligence:

    When America.gov did not know something, it generally said so.

    A New Front Door With 29,000 Rooms

    The General Services Administration unveiled America.gov on September 29 as what it calls a new digital “front door” to federal information and services.

    According to GSA’s launch announcement, America.gov combines information from more than 29,000 government websites with AI-enhanced search technology. Instead of navigating the federal government’s sprawling collection of agency websites, GSA says Americans will eventually be able to use a single interface to find information and conduct transactions.

    The White House goes further.

    Its September 29 fact sheet says Americans can currently “ask any question” and receive an up-to-date answer. Eventually, the administration says, people will be able to complete transactions including passport renewals and Medicare enrollment without being dispatched back into what the White House rather accurately calls a “maze of websites.”

    President Donald Trump also signed an executive order directing GSA, the National Design Studio and the Office of Management and Budget to establish America.gov as the central entry point for covered federal services.

    The government, in other words, has created a website designed to spare Americans from having to understand how the government created its websites.

    There is a certain elegance to that.

    There is also an immediate accountability question.

    What, exactly, are those 29,000 websites?

    GSA’s announcement does not provide the inventory or explain what it counts as a website. A domain could be a website. So could a subdomain. An agency might operate multiple applications or collections beneath one domain. And saying that America.gov “combines information” from 29,000 websites does not tell us whether every document hosted on those sites is searchable by the system.

    As our experiment would demonstrate, that distinction matters.

    Bureaucracy Times sent GSA a media inquiry Tuesday night asking for the inventory, the definition used to arrive at the 29,000 figure, the inclusion criteria, update frequency, source-ranking methodology and information about the AI models and providers operating the system.

    We also asked about records retention, system instructions, configuration changes and whether America.gov was designed and tested for research uses such as locating FOIA offices and federal records.

    The inquiry was sent at 9:50 p.m. Eastern. We will update this article if GSA responds.

    America.gov Has Been Here Before

    The name is new to many Americans. The domain is not.

    America.gov previously belonged to a State Department public-diplomacy operation intended largely for international audiences.

    A 2011 Senate Foreign Relations Committee record states that the old America.gov was transitioned to the State Department’s IIP Digital operation and archived on April 1, 2011.

    It was not an obscure little government webpage when it disappeared.

    According to the Senate record, America.gov had averaged approximately 1.67 million monthly page views and 925,288 monthly visitors, with 63.52 percent of visits coming from outside the United States.

    The federal government has therefore resurrected the address of what was once an American public-diplomacy operation and turned it into an AI gateway for Americans trying to navigate their own government.

    Washington does recycle.

    Sometimes it even recycles websites.

    Our Test Was Not Random

    We wanted to know whether America.gov could do more than tell someone where to renew a passport.

    Could a journalist use it?

    Could a researcher?

    Could an ordinary citizen trying to understand an inspector general report, federal contract or regulatory requirement use the government’s new front door to find information scattered through the federal bureaucracy?

    Rather than invent trivia questions, we selected 15 questions from Bureaucracy Times reporting and research conducted during the past month.

    That gave us something unusually useful when testing an AI system: a control.

    For some questions, we already knew the answer from federal records.

    For others, our reporting had established that the answer was not publicly available.

    In several cases, we possessed agency correspondence America.gov should not have access to at all.

    We recorded the responses and compared them against our existing reporting. Answers were classified as correct, partially correct, incorrect, unsupported or no answer. A separate category was created when America.gov’s safeguards prevented a question from being submitted.

    The final score was nine correct and six partially correct.

    There were no answers we classified as incorrect or unsupported.

    There were no hallucinations we could identify.

    That last finding deserves more attention than the score.

    When America.gov Couldn’t Find the Answer

    The system’s weaknesses appeared quickly.

    We asked how many separated Customs and Border Protection contractors retained system access after separation and how many contractor separation records were delayed more than 30 days.

    America.gov located the correct 2026 Department of Homeland Security inspector general report.

    Then it stopped.

    It said the available material did not provide the requested counts and suggested opening the full OIG report.

    The numbers are 1,208 and 623.

    America.gov had found the building but apparently couldn’t get into the room where the numbers were kept.

    A similar problem emerged when we asked how much the federal government has paid from the Judgment Fund because the Department of Energy failed to begin accepting commercial spent nuclear fuel in 1998.

    America.gov reconstructed the history accurately. It found totals of approximately $3.7 billion through fiscal 2013, $7.4 billion through fiscal 2018 and $10.1 billion as of September 30, 2022.

    It correctly distinguished those payments from DOE’s projected future liability.

    But our reporting had already located a newer figure of approximately $12.2 billion.

    America.gov’s answer was accurate as far as it went. It simply did not go far enough.

    The same happened when we asked how much commercial spent nuclear fuel is currently stored around the country. America.gov found official figures ranging from roughly 86,000 to more than 90,000 metric tons and correctly explained that there is no operating permanent federal repository.

    Newer federal information puts the inventory above 95,000 metric tons.

    Again, the problem was not fabrication.

    It was freshness.

    Then It Found the Nickel

    We turned up the difficulty.

    DOE has spent decades dealing with thousands of tons of radioactively contaminated nickel at the Paducah Gaseous Diffusion Plant in Kentucky.

    Bureaucracy Times had previously identified a curious discrepancy. An older DOE environmental assessment described 8,500 contaminated nickel ingots weighing approximately 9,350 short tons. Later DOE documents described approximately 9,700 tons.

    We asked America.gov for both figures.

    It found them.

    More impressively, the system corrected a chronological assumption embedded in our question rather than obediently accepting it. It explained that the earlier environmental assessment was the contemporaneous source and then located the later 9,700-ton figure.

    America.gov went one better.

    It found that DOE was still citing approximately 9,700 tons of contaminated nickel in a November 2025 Expression of Interest.

    That was a genuine research result.

    For a system less than a day old, searching through decades of obscure nuclear cleanup documents, it was impressive.

    Then we gave it a federal contract number.

    And America.gov decided we had gone too far.

    Please Remove the Government’s Personal Information

    We asked:

    How much recyclable-material credit did DOE receive under Contract DE-AC05-97OR22576 at the East Tennessee Technology Park?

    America.gov would not allow the question to be submitted.

    The contract number, DE-AC05-97OR22576, was highlighted in red.

    The interface instructed us: “Remove personal information before sending.”

    We preserved a screenshot.

    To be precise, our test establishes only that this particular public federal contract identifier triggered America.gov’s personal-information safeguard. It does not establish that all federal contract numbers will do so.

    But federal research depends heavily on identifiers. Contract numbers, solicitation numbers, docket numbers, case numbers, grant numbers, award identifiers and FOIA tracking numbers are the connective tissue of the government’s public record.

    A research system that occasionally mistakes those identifiers for private information could have a meaningful limitation.

    So we removed the contract number and tried again.

    America.gov immediately found the correct BNFL contract.

    It identified the original $238 million fixed-price agreement, approximately 129,000 tons of potentially reusable metal, $12 million in approved upward price adjustments and a roughly $40 million claim associated with DOE’s moratorium on releasing recycled nickel.

    It could not find the figure we wanted.

    Our previous research identified a recyclable-material credit of $55,569,748 in contract documentation.

    America.gov had once again arrived remarkably close to the answer without retrieving the particular fact buried in the record.

    The Questions With No Answers

    Then something interesting happened.

    America.gov began performing better when we asked questions that did not have public answers.

    We asked how much DOE has spent storing the contaminated Paducah nickel since the proposed sale failed decades ago.

    We do not know.

    America.gov did not know either.

    “I cannot invent that number,” it responded.

    That sentence may be the most important thing America.gov told us all night.

    We asked how many Form 10 attrition events involving frozen pension plans the Pension Benefit Guaranty Corporation receives each year.

    PBGC recently waived certain reporting requirements after saying its review showed such events rarely warranted additional scrutiny, but the agency did not publish the annual count underlying that conclusion. Bureaucracy Times sent PBGC a media inquiry seeking precisely that data earlier this month.

    America.gov found the technical update, explained the reporting requirements and correctly concluded that the annual figure could not be established from the published information.

    It did not grab some other PBGC statistic and force it into the hole.

    Then we asked whether federal renewable-fuel regulations require producers to prove that used cooking oil was lawfully obtained from the restaurant or original owner before generating Renewable Identification Numbers.

    America.gov correctly distinguished EPA’s feedstock documentation requirements from proof of lawful ownership.

    We followed with an even nastier question: Can stolen used cooking oil qualify for the Section 45Z Clean Fuel Production Credit if a fuel producer unknowingly purchases it?

    Bureaucracy Times has already asked Treasury and the IRS essentially that question.

    We have not found a definitive public answer.

    Neither did America.gov.

    It explained what Section 45Z requires, identified the relevant recordkeeping requirements and then stopped.

    It did not convert the absence of a prohibition into eligibility.

    It did not convert the absence of an eligibility provision into prohibition.

    That is exactly what an information system should do.

    It Also Knew What It Shouldn’t Know

    One of our tests involved Zachary Hess Dawson, who pleaded guilty in a federal prosecution involving more than $350,000 in fraudulently obtained VA benefits.

    Bureaucracy Times previously submitted a FOIA request to the VA Office of Inspector General seeking records concerning how the alleged fraud was detected and how the benefits continued.

    VA OIG denied our request under FOIA Exemption 7(A), citing the pending criminal proceeding.

    We asked America.gov why VA OIG denied access.

    It couldn’t answer.

    That was the correct response.

    The denial letter was sent directly to us. America.gov found the prosecution and the OIG investigation, but it could not find a public agency determination explaining our particular denial.

    Rather than infer that the pending prosecution was probably the reason, America.gov said it could not state why the request was denied.

    We then asked how VA continued paying Dawson from 2017 through 2025 and whether his arrest or incarceration triggered discovery of the alleged fraud.

    Again, America.gov found the known facts and refused to construct a causal story from them.

    The public record does not answer those questions.

    Neither did the machine.

    For journalists accustomed to watching generative AI confidently connect two dots with a third dot it just invented, restraint is worth noticing.

    The Classified Clocks Exposed Another Weakness

    Not every miss involved an obscure historical document.

    Earlier this month, the Department of War inspector general released an audit concerning security controls over Defense Regional Clocks.

    All 10 recommendations are classified.

    We asked America.gov why.

    It correctly determined that the government has not publicly disclosed the underlying classification rationale. It also found the original project announcement showing that the audit began after Hotline allegations.

    But America.gov said it could not verify publicly that there were 10 recommendations.

    That information is sitting on another government website.

    Oversight.gov publicly lists the report and identifies 10 recommendations, each classified.

    Here, America.gov successfully searched one portion of the federal information ecosystem while missing another federal source containing the information it lacked.

    That is an important limitation for a service advertised as combining information from more than 29,000 government websites.

    What does “combines” mean?

    Our test suggests it does not necessarily mean that every relevant piece of information across those sites will be retrieved and synthesized for every query.

    Public Records Still Matter

    We saw the same boundary when we asked whether Virginia election officials had provided voter-list products to organizations involved in an overseas-voter mail inquiry.

    America.gov correctly explained Virginia’s restrictions on distributing voter information and said it could not verify that such a transfer occurred.

    We knew something it didn’t.

    Virginia’s Department of Elections had already responded to our records request saying it found no responsive records for the requested voter-list products involving the specified organizations.

    That isn’t a failure by America.gov.

    It is a reminder of what the system cannot replace.

    Search can tell a journalist what the government has published.

    A records request can establish what the government has not.

    America.gov may make federal research considerably easier. It does not eliminate the need to ask agencies questions they haven’t already answered on the internet.

    The bureaucracy lives to see another FOIA.

    The MMTLP Test

    Our fourteenth question concerned MMTLP, the ticker symbol at the center of a years-long dispute involving FINRA’s December 2022 trading halt.

    America.gov accurately reconstructed FINRA’s stated reason for imposing a U3 extraordinary-event halt, including the settlement problem surrounding the distribution of Next Bridge Hydrocarbons shares.

    But it failed to locate a separate systems coding error involving MMTLP that FINRA subsequently acknowledged and that Bureaucracy Times had already identified during earlier reporting.

    Again, the prominent portion of the record was easy to find.

    The less prominent official disclosure was not.

    America.gov’s source discipline nevertheless held. It encountered investor allegations concerning other supposed errors and refused to characterize those allegations as official FINRA findings.

    Another partial answer.

    Another refusal to manufacture the missing piece.

    One Last Trap

    Our final question concerned the U.S. Arctic Research Commission.

    We asked for aggregate payroll and employer-paid personnel costs processed by the Interior Business Center for fiscal years 2024 and 2025.

    We don’t have those numbers yet. They are part of an ongoing records inquiry.

    America.gov found something that looked tantalizingly close: NSF budget tables showing USARC personnel compensation and benefits.

    For fiscal 2024, it found approximately $1.008 million.

    A less careful system could have handed us that number.

    America.gov didn’t.

    It explicitly distinguished budgeted personnel compensation and benefits from actual payroll and employer-paid costs processed through IBC.

    “I cannot confirm those IBC figures from these official sources,” it said.

    Correct.

    America.gov Wins Round One

    This was not a comprehensive technical evaluation.

    Fifteen questions cannot establish the overall accuracy of an AI system drawing information from a federal web environment that GSA says encompasses more than 29,000 websites. Our questions were also deliberately unusual. They came from the kinds of federal audits, contracts, regulatory disputes and records inquiries Bureaucracy Times spends its time digging through.

    But that was the point.

    We weren’t interested in whether America.gov could tell us how to apply for Social Security.

    We wanted to know what happened when a citizen walked through the government’s shiny new front door carrying a shovel.

    The answer, after one evening of testing, is encouraging.

    America.gov was not omniscient. It missed current figures. It sometimes failed to retrieve facts from reports it had correctly identified. It failed on some cross-government searches. And its privacy safeguard apparently looked at a Department of Energy contract number and decided we were trying to disclose someone’s personal information.

    There is work to do.

    But the failure we expected to find never arrived.

    We could not get America.gov to make something up.

    Several of our hardest questions were hard precisely because the answers remain unknown. America.gov consistently recognized that distinction. It found related evidence, explained what that evidence established and stopped before turning inference into fact.

    For researchers and journalists, that characteristic may ultimately matter more than whether every answer is immediately complete.

    Completeness can be investigated.

    A confident fabrication can send the investigation in the wrong direction entirely.

    There are also larger accountability questions still unanswered. GSA says America.gov draws from more than 29,000 government websites, but we have asked for the inventory behind that claim. We want to know how sources are selected and ranked, how quickly information is updated, which AI models are involved, how configuration changes are documented and what records are preserved when the system’s behavior changes.

    Those questions became particularly relevant on launch day. The Associated Press reported that America.gov initially answered several politically sensitive questions using government information, then began refusing some of those questions while President Trump was still speaking at the launch event. That does not tell us why the behavior changed or who changed it. It does demonstrate why configuration histories and change logs for a federal information system matter.

    GSA itself maintains an agency-wide AI directive requiring risk management, transparency and lifecycle accountability for covered AI systems. America.gov is now a very public test of what those principles mean when artificial intelligence becomes an intermediary between citizens and their government.

    And this is only the beginning.

    The White House says America.gov will eventually move beyond answering questions and begin conducting government transactions. Passport renewals and Medicare enrollment are among the services promised.

    At that point, accuracy will become something more than a search-quality metric.

    For now, though, America.gov has been online for one day.

    We spent its first evening feeding it questions about radioactive nickel, stolen cooking grease, pension reporting, classified clocks, securities regulation, federal payrolls, benefits fraud and a contract number it apparently wished we had kept to ourselves.

    It survived.

    More importantly, when it didn’t know the answer, it usually had the good sense to say so.

    For the federal government’s new AI front door, that is a surprisingly good place to start.

    Fediverse reactions

  • The FAA Knows Someone Could Fake Its Air Traffic Instructions. It’s Still Working On That.

    Somewhere within range of a major airport, a person with a software-defined radio that costs less than a nice dinner could transmit a message that looks, to a pilot’s instruments, exactly like a legitimate clearance from air traffic control. There is currently no reliable way for that pilot to know the difference. This is not the premise of a thriller; it is the finding of a Government Accountability Office report released September 21, 2026, on the cybersecurity of the systems that keep more than 44,000 daily flights and 3 million passengers from colliding. The FAA has known about this class of problem for years. It has, by GAO’s account, mostly gotten around to writing memos about it.

    The ghosts in the frequency

    The National Airspace System runs on radio frequency signals, the same physical medium used for GPS, radar, and every other invisible thread holding modern aviation together. GAO catalogs three ways to attack it (interference, spoofing, and jamming), each capable of degrading, faking, or simply killing the signals aircraft and controllers depend on. None of this is theoretical. In 2012, an individual jamming GPS transmissions near Newark Airport degraded a landing-approach system during pre-deployment testing. In 2022, GPS interference with signs of spoofing near a Dallas airport caused controller confusion and more than 230 departure delays; the same year, an unauthorized transmitter near Denver International disrupted navigation accuracy for civil flights. GAO also notes, in the driest possible register, that a December 2024 Azerbaijan Airlines flight was shot down amid GPS disruptions linked to the Russia-Ukraine conflict, an approach GAO describes as having “become more complex than usual” due to pilot and controller miscommunications. International GNSS interference is now concentrated enough to have hotspots, mapped in the report like weather systems, radiating out of the Middle East, Eastern Europe, Southeast Asia, and the Baltic region.

    FAA is well aware of all this. Awareness, GAO’s report makes clear, is where the agency’s competence currently ends. This is also the third GAO aviation cybersecurity report in as many years; a September 2024 report found 51 of FAA’s 138 air traffic control systems outright unsustainable, and a July 2026 report found FAA had fully implemented only two of its own seven cybersecurity strategy goals.

    Assessed, in a general sort of way

    Federal information security law requires agencies to run risk assessments for individual systems, evaluating specific threats, vulnerabilities, and likely impact under NIST Special Publication 800-53 Rev. 5, and FAA’s own internal order requires the same. For seven of the eight spectrum-dependent NAS systems GAO reviewed, FAA had not done this. Instead it points to one general spectrum risk assessment covering everything at once, which GAO politely observes “does not satisfy the NIST recommendation” because it treats every system’s exposure as roughly identical rather than evaluating each one’s actual architecture and controls.

    The paperwork underneath these systems is not in better shape. Four of the eight still cite a NIST standard that was superseded in 2021. Five are missing required security controls for high-impact systems, including contingency planning and denial-of-service protection. One system’s documents cannot agree with each other on whether the system is high impact or merely moderate. One had not had its required annual security assessment done since 2023. A separate DOT Inspector General audit, cited in the report, found the same pattern of stale NIST standards and missing controls across FAA’s high-impact systems.

    Real-time monitoring, the thing that would actually let FAA notice an attack while it is happening rather than reconstructing it afterward, does not fully exist. FAA’s Spectrum Engineering office told GAO it has no 24/7 monitoring capability and lacks the specialized tools to detect interference or spoofing as it occurs; instead it investigates after someone reports a problem. The agency’s own Wide Area Augmentation System can flag degraded GPS performance but cannot detect spoofing or identify where it came from. FAA officials cited funding constraints as the reason certain monitoring tools have not been deployed to high-risk sites. It is worth sitting with that sentence; the nation’s air traffic infrastructure lacks live detection for signal spoofing because of a budget line item.

    A coalition of the partially willing

    FAA does talk to other agencies about all this, and GAO gives it credit for two of the eight leading interagency collaboration practices it measured against, defining common outcomes and leveraging shared resources, chiefly through the Aviation Cyber Initiative, the Purposeful Interference Response Team, and the Interdepartmental Radio Advisory Committee. The other six practices were only partially met, and the gap has a consistent shape. Inside the formal interagency groups, FAA has charters, bylaws, and defined roles. Outside them, it has almost nothing. FCC and DOD officials told GAO they routinely share information with FAA through informal channels with no documented policy governing how or when. Some aviation industry stakeholders told GAO they were unsure which FAA office to even contact about a spectrum issue, and some reported wanting more access to the incident coordination calls that already exist. FCC itself was apparently dropped from continued participation in the Aviation Cyber Initiative after an earlier interagency taskforce wound down in 2021, and nobody seems to have noticed enough to fix it.

    Clearance to land, or so says somebody

    The most concrete vulnerability in the report involves two applications that pilots and controllers actually use every day, ACARS, which handles operational messaging between aircraft and airlines, and CPDLC, which lets controllers send digital text clearances as an alternative to voice. Both predate modern cryptography as a design assumption. Neither is encrypted. Neither authenticates the sender. GAO walks through, with illustrated step-by-step diagrams, how a malicious actor could pull public flight-tracking data, transmit a spoofed clearance cancellation over VHF using nothing more exotic than a software-defined radio, and force a pilot into a verbal shouting match with real air traffic control to sort out which instruction is genuine. Do that to several aircraft at once and you have delayed departures, an overloaded controller workload, and, in a worse version of the same scenario, a pilot who follows a fabricated altitude or route change believing it came from ATC.

    FAA’s defense against all this is essentially procedural rather than technical: message-format standardization, mandatory acknowledgments, and the expectation that a suspicious pilot will simply pick up a voice radio and ask. These are reasonable stopgaps. They are also, as GAO notes, “not designed to provide the same level of protection as modern cryptographic controls,” which is a generous way of saying the agency is trusting human vigilance to compensate for a protocol built before anyone thought to secure it. A next-generation replacement, built on standard internet protocols, is in testing at select locations. FAA has not given GAO a timeline for when, or whether, it will actually replace the systems currently in use.

    GAO issued nine recommendations, covering system-specific risk assessments, consistent categorization, real-time monitoring, interagency accountability, sustained leadership, formal information-sharing agreements outside the existing clubs, broader stakeholder inclusion, and a plan to actually authenticate ACARS and CPDLC traffic. The Department of Transportation, replying on FAA’s behalf, concurred with all nine and promised a detailed response within 180 days, the standard federal-agency interval for “we will get back to you, eventually, with something.” Bureaucracy Times will be checking the calendar.

  • Who Decides When a Quantum Computer Actually Works?

    The Quantum Genesis Files, Part Four

    Washington has spent years funding quantum computers, testing them and listening to companies explain what their machines can do. Now DOE is building an apparatus to decide when one of them has actually delivered.

    Building a fault-tolerant quantum computer is difficult.

    Determining that someone has built one may prove surprisingly complicated too.

    That problem sits quietly underneath the Department of Energy’s new Quantum Genesis Q Competition.

    DOE wants private companies to demonstrate quantum computers with at least 100 logical qubits, hundreds of millions of fault-tolerant operations and scientific applications worthy of the Department’s research mission.

    Reach those milestones and substantial federal money becomes available.

    But first somebody has to verify the milestones.

    That means determining whether logical qubits are really behaving logically, whether error rates are low enough, whether operations can be repeated reliably and whether a supposedly useful scientific computation actually represents something conventional computers cannot reasonably accomplish.

    DOE has decided not to leave those judgments entirely to the companies building the machines.

    The Department is building referees.

    The $45 Million Referee

    When DOE announced the Quantum Genesis Q Competition on September 17, the $215 million competition naturally received most of the attention.

    Sitting beside it was another announcement.

    DOE plans to spend up to $45 million establishing a Quantum High-Performance Computing Validation and Verification Testbed through its National Laboratories.

    Fourteen million dollars is available from fiscal year 2026 funding. The remaining planned funding depends upon future congressional appropriations.

    Its assignment is remarkably broad.

    DOE says the testbed must be capable of characterizing and validating the entire quantum-computing stack.

    That includes physical hardware.

    Quantum gates.

    Logical architectures.

    Algorithms and applications.

    Classical control systems.

    The testbed will support the Q Competition, meaning companies seeking milestone payments will face something more substantial than a review of specifications supplied by their own engineers.

    DOE wants independent evidence.

    The Department is effectively spending money to answer a question that has followed quantum computing almost from its beginning.

    What exactly should count as success?

    Qubits Make Terrible Scoreboards

    Traditional computers are comparatively easy to describe.

    Processor speeds, memory, storage and standardized benchmark tests can provide imperfect but reasonably understandable comparisons.

    Quantum computers are another matter.

    Two machines can contain similar numbers of physical qubits while having dramatically different capabilities.

    One architecture may produce excellent individual gate fidelity while struggling with connectivity.

    Another may possess more qubits but suffer from errors that limit useful circuit depth.

    A machine may perform impressively on one specialized benchmark and considerably less impressively on another.

    Even the same quantum processor can produce different results depending upon calibration, software, circuit compilation and the expertise of the person operating it.

    That makes simple qubit counts a particularly poor substitute for useful performance.

    The federal government has known this for years.

    And some of its laboratories have been trying to solve the measurement problem long before Quantum Genesis existed.

    Sandia Has Been Building Yardsticks

    At Sandia National Laboratories, the Quantum Performance Laboratory has made quantum-computer assessment its business.

    The laboratory develops methods for measuring how quantum hardware actually performs.

    Its researchers study failure mechanisms, devise performance metrics, build predictive models and create protocols for testing experimental processors.

    They also develop and maintain pyGSTi, an open-source software package used to characterize quantum processors.

    Most importantly for Quantum Genesis, Sandia says the laboratory provides quantum-hardware assessment capabilities directly to DOE and the federal government.

    That makes Sandia’s work particularly relevant to the new competition.

    The Quantum Performance Laboratory has been publishing research on subjects such as randomized benchmarking, gate-set tomography, application-oriented benchmarks and fault-tolerance thresholds for years.

    Those terms may sound painfully specialized.

    Their purpose is relatively straightforward.

    A manufacturer can say a quantum processor performs at a certain level.

    Someone else needs a scientifically defensible way to check.

    Sandia has spent years developing the ruler.

    Oak Ridge Tried Measuring the Machines

    Oak Ridge National Laboratory approached the problem from another direction.

    Its Quantum Computing User Program has provided researchers access to commercial quantum computers from multiple companies.

    That arrangement allowed scientists to do something especially useful.

    They could compare machines.

    In 2023, Oak Ridge researchers reported what the laboratory described as the first independent comparison of leading quantum computers.

    The team examined 24 quantum processors.

    They then compared their results against performance figures promoted by vendors including IBM, Rigetti and Quantinuum.

    The exercise demonstrated precisely why verification matters.

    Most machines performed reasonably close to the vendors’ reported capabilities.

    But researchers had difficulty reproducing the highest advertised performance figures consistently.

    That did not mean the vendors were lying.

    Oak Ridge’s researchers explicitly cautioned against that conclusion.

    Manufacturers know their machines extraordinarily well and can optimize their hardware in ways ordinary users may not reproduce.

    That distinction may sound minor.

    For Quantum Genesis, it is enormous.

    DOE is not paying for the performance a manufacturer’s best engineers can coax from a carefully optimized demonstration.

    It is trying to determine whether a machine represents a scientifically useful computing capability.

    Those are not necessarily the same standard.

    DARPA Is Asking an Even Harder Question

    The Defense Advanced Research Projects Agency has developed its own verification system.

    DARPA’s Quantum Benchmarking Initiative is trying to determine whether any current quantum-computing approach can reach what the agency calls utility-scale operation by 2033.

    Its definition is intentionally economic as well as technical.

    The computational value of the machine must exceed its cost.

    Companies move through three stages.

    Stage A asks whether the company has a plausible concept.

    Stage B requires a detailed research and development plan, identification of technical risks and prototypes designed to reduce those risks.

    Then comes Stage C.

    At that point, a government verification-and-validation team determines whether the proposed utility-scale computer can actually be constructed as designed and operated as intended.

    DARPA describes the program as independent third-party verification.

    That distinction matters.

    The agency is not trying to crown the best quantum company.

    It says explicitly that the Quantum Benchmarking Initiative is not a competition intended to narrow the industry to a predetermined group of winners.

    Multiple approaches could succeed.

    One could succeed.

    None could succeed.

    The job of the government evaluators is to determine which claims survive examination.

    Two Federal Finish Lines

    DARPA and DOE are therefore examining related questions with different finish lines.

    DARPA is asking whether a company can build an industrially useful quantum computer by 2033.

    DOE wants scientifically relevant fault-tolerant quantum computing beginning around 2028.

    DOE’s target is tied to scientific utility.

    DARPA’s is tied to utility at industrial scale.

    The distinction creates an interesting federal experiment.

    A company could theoretically demonstrate a machine valuable for certain DOE scientific applications before proving that the same technological architecture can become economically useful at industrial scale.

    That makes Quantum Genesis something other than a miniature version of DARPA’s program.

    The agencies are measuring different stages in the evolution of the technology.

    They are also developing independent government expertise rather than accepting industry roadmaps at face value.

    The Government Is Learning How to Say No

    That may be the least glamorous but most important part of the entire Quantum Genesis story.

    Federal quantum policy has spent years asking how government can accelerate the technology.

    Fund research.

    Create national centers.

    Connect companies with laboratories.

    Purchase commercial hardware access.

    Support domestic manufacturing.

    Develop algorithms.

    Train scientists.

    Those activities help technology move forward.

    Verification serves a different purpose.

    It gives the government the ability to say no.

    No, that qubit count does not demonstrate useful performance.

    No, that benchmark does not prove fault tolerance.

    No, that scientific calculation does not yet represent meaningful advantage.

    No, that roadmap has not sufficiently addressed the engineering risks.

    Or eventually, perhaps, yes.

    Without an independent ability to make those judgments, a government quantum program risks becoming dependent upon the companies seeking its money to explain whether their own machines deserve it.

    DOE appears determined to avoid that arrangement.

    From Laboratory Instrument to Federal Infrastructure

    The Department’s ambitions extend beyond the current competition.

    Its 2026 roadmap envisions quantum processors eventually becoming part of DOE’s broader scientific-computing infrastructure.

    That means quantum machines would operate alongside conventional high-performance computers rather than replacing them.

    Researchers could move portions of scientific workloads onto quantum processors when those processors offer an advantage.

    DOE ultimately envisions a dedicated Quantum Computing User Facility built around that hybrid model.

    That changes the stakes of verification.

    The Department is not merely evaluating prototypes for a technology prize.

    It is potentially deciding what equipment eventually belongs inside the federal scientific-computing enterprise.

    DOE already operates some of the most powerful conventional supercomputers in the world.

    If quantum processors are going to become another instrument in that infrastructure, the Department needs to know not merely whether they work.

    It needs to know when they are useful.

    The Bureaucracy Behind the Breakthrough

    Quantum computing is often presented through images of exotic hardware, gold-colored cryogenic equipment and promises about calculations that would take conventional supercomputers longer than the age of the universe.

    The less photogenic part of the story involves measurement protocols, error models, validation plans, benchmark suites and federal scientists asking manufacturers to run the test again.

    That may ultimately matter just as much.

    The Quantum Genesis Q Competition is attempting to push private companies toward a fault-tolerant machine.

    The new DOE validation testbed will try to determine whether they get there.

    Sandia has spent years developing quantum-performance measurements.

    Oak Ridge has spent years putting commercial processors in front of independent scientific users.

    DARPA has built a staged process for dissecting company roadmaps and eventually verifying complete systems.

    Different agencies and laboratories are approaching the problem from different directions.

    But they are converging on the same principle.

    Extraordinary technological claims require considerably more than extraordinary press releases.

    They require somebody with a ruler.

    The Race Begins

    That brings the first four Quantum Genesis Files full circle.

    Part One followed the decade of DOE research that preceded the new $215 million competition.

    Part Two followed the federal money that helped construct the broader quantum ecosystem.

    Part Three mapped many of the companies and architectures already familiar to federal laboratories and evaluators.

    And Part Four arrives at the people responsible for deciding whether any of it actually works.

    The next major event is October 19, when applications for the Quantum Genesis Q Competition are due.

    Until then, we do not know who will enter.

    We do know considerably more about the track on which they will run.

    The federal government helped finance portions of the underlying science.

    Its laboratories helped test the machines.

    Its agencies are supporting manufacturing.

    Its researchers developed benchmarks.

    Its scientists are building verification tools.

    And now DOE has attached money to the finish line.

    For a technology famous for existing in multiple states at once, Quantum Genesis has reached at least one surprisingly definite condition.

    The government is no longer merely asking whether useful fault-tolerant quantum computing is possible.

    It is preparing to measure it.

  • Two Minutes on the House Floor, Five Days on the Clock

    The most consequential item in Monday’s Congressional Record is a piece of mail. The White House’s letter proposing 11 rescissions of budget authority, dated Friday, was logged and referred to committee. The House then convened at noon, adjourned two minutes later, and set its next session for Thursday, October 1. The money in question expires on September 30.

    The Arithmetic of Rounding Up

    The package totals $809.5 million, which Democrats on the Senate floor kept rounding up to nearly a billion. The largest slice is $567.4 million from HHS refugee and entrant assistance, followed by $69.6 million of a $70.3 million international education line and $56.1 million from HUD housing counseling. It also claws back $15 million from the Justice Department’s Community Relations Service, an office the administration eliminated in late 2025, which makes this the rare rescission that arrives after the funeral.

    A Calendar Doing the Work of a Statute

    The device is a pocket rescission. The Impoundment Control Act gives Congress 45 days to act on a request, and the administration reads that as permission to sit on the funds until they lapse. The Government Accountability Office concluded in 2018 and again in 2025 that the law does not allow this. The Supreme Court last September let the administration keep withholding roughly $4 billion while litigation continued, but said its order was not a ruling on the merits. Susan Collins, the Republican who chairs Senate Appropriations, called this one unlawful too. Nothing is settled, and the calendar does the rest.

    The Senate Stayed Until Nearly Eleven at Night

    The Senate spent its evening passing the Protect College Sports Act, 77 to 22, with 50 Republicans and 27 Democrats in favor. A Booker amendment on catastrophic injury insurance for athletes failed 49 to 50. By unanimous consent, senators also passed the House’s penny bill (H.R. 10167), which ends circulating minting and rounds cash sales to the nearest five cents, plus a terrorism risk insurance reauthorization (S. 4395) and a bill (H.R. 997) letting the National Taxpayer Advocate hire its own counsel.

    Two inspector general nominations arrived, Mark Priebe for Education and Robert Steinau for NASA, along with Jeffrey Zisk for the Postal Regulatory Commission. Majority Leader Thune filed cloture on the Stop Insider Trading Act, which would restrict stock holdings for members of Congress and their families, then withdrew the underlying motion; the Daily Digest schedules the cloture vote for Wednesday.

    Congress spent Monday retiring the penny and studying the price of college football, while eleven appropriations slipped quietly into a folder. It is a fine way to cancel money. Simply schedule the vote for after it is gone.

    Fediverse reactions

  • The Design Standard NNSA Lowered to Keep Its Plutonium Factory on Schedule

    Of the five construction projects that the National Nuclear Security Administration has ever bothered to isolate into their own contract line items, the Savannah River Plutonium Processing Facility is the newest and the biggest. According to a new GAO report on how NNSA uses these line items, the facility got its own cost-plus-award-fee line in 2023, its own tailored performance evaluation, and, according to officials, a fee plan that took a group of thirty to forty staff and contractors to put together. The theory, as always, was that isolating the deliverable would isolate the accountability.

    What GAO’s own separate assessment of NNSA’s major projects found, published the same month, is that the agency has spent the last two years quietly lowering the bar the project has to clear rather than raising the pressure on the contractor to clear it.

    The Standard NNSA Lowered for Itself

    In October 2024, NNSA adopted what it calls a second revised tailoring strategy for the plutonium facility’s Main Process Building, according to GAO’s February 2026 assessment of NNSA’s major projects. The strategy folded the scope of three other projects into the Main Process Building, and it did something considerably more interesting to the government’s own rulebook. It let NNSA establish a formal cost and schedule baseline once the design was only sixty percent finished, instead of the ninety percent DOE’s own project management order otherwise requires. It also let construction begin before that baseline existed at all.

    NNSA approved ten early procurement and site preparation packages under this looser standard, intended to get lower risk work started while the design was still being finalized. According to GAO, the agency could not fully execute even that plan, because the contractor kept falling behind on delivering the design packages that were supposed to accompany the work. Officials told GAO they are still struggling to hire qualified design engineers and that the contractor continues to miss performance targets. The project’s final design review, originally set for March 2026, has slipped to September 2026. When the contractor submitted its package for an independent cost estimate in May 2025, GAO found it was insufficient to actually conduct the estimate; NNSA is waiting on a resubmission in early 2026, at which point officials hope the design will finally be close to the ninety percent maturity the agency waived in the first place.

    The cost, meanwhile, has done what cost usually does at Savannah River. NNSA’s own fiscal year 2026 budget justification puts the Main Process Building at over twenty two billion dollars, against a preliminary range of $6.9 billion to $11.1 billion that NNSA itself published at the project’s first major milestone. Officials told GAO the September 2035 completion date, already the later end of that earlier range, may slip further. A revised cost and schedule estimate is not due for internal NNSA approval until September 2026.

    A Building With a Prior Conviction

    None of this is Savannah River’s first attempt at doing something complicated with plutonium in that same footprint. The plutonium facility is being built inside and around the shell of the site’s Mixed Oxide Fuel Fabrication Facility, a program NNSA spent roughly fifteen years and more than eight billion dollars on before mothballing it in 2014. That project, too, started with a modest number. DOE estimated MOX construction at $1.6 billion in 2004. By 2013, GAO was confirming that NNSA had quietly added another two billion dollars to the estimate even as the facility was supposedly nearing completion, after years of the agency declining to disclose a rebaselined figure publicly. The facility never processed a gram of commercial fuel. NNSA converted the shell into the plutonium pit factory it operates today.

    There is a version of this story where the agency learned something from the MOX years and applied it to the pit factory. The tailoring strategy GAO documented in the new report suggests the lesson NNSA actually absorbed was narrower: that a hard design-maturity threshold is an obstacle to be waived when the schedule is inconvenient, not a floor to be held.

    Now Firing the Contractor, Unrelatedly

    Whatever the line item and its fee plan were supposed to buy NNSA, it apparently was not enough confidence in the incumbent. Savannah River Nuclear Solutions, the Fluor led joint venture that has run the site since 2008 and is building the plutonium facility, received a project management rating of less than adequate in its fiscal year 2024 performance evaluation, according to a review DOE released in February 2025, which cited the absence of a credible overall project schedule. In February 2026, NNSA released a request for proposals for a standalone follow-on management and operating contract for the entire Savannah River Site, worth an estimated sixty one and a half billion dollars over a potential twenty years and four months, a competition NNSA’s own announcement frames as separate from the plutonium project specifically but that plainly reflects an agency no longer content with how its current contractor performs. NNSA extended the incumbent’s contract through September 2027, worth an additional $2.3 billion, explicitly to preserve workforce continuity while the recompete plays out, which is its own small admission that the agency does not trust the transition to go smoothly either.

    The new M&O contract, whoever wins it, will not just run the plutonium factory. It also covers the storing, processing, and long term management of the nuclear materials and waste that Savannah River exists to handle, all under the same undocumented, case by case logic GAO’s line item report says NNSA still uses to decide which pieces of that work get isolated into their own accountability structure and which pieces stay folded into the general M&O scope where waived standards and quiet rebaselines are harder to see. A cost-plus-award-fee line item and a thirty person fee plan gave GAO something to point to when NNSA needed to demonstrate it was taking the plutonium facility seriously. It did not stop the agency from lowering its own design standard to keep pouring concrete, and it did not stop the government from concluding, two years after handing the project its own line, that the company running it needs to be replaced.

    Sources: GAO, NNSA Contract Line Items: Agency Should Document Criteria for Use and Capture Lessons Learned (GAO-26-108409, September 2026); GAO, NNSA Construction Projects: Cost and Schedule Performance (GAO-26-107777, February 2026); GAO, Mixed Oxide Fuel Fabrication Facility: DOE Needs to Provide More Consistent and Timely Cost Information (GAO-13-283, 2013). Related: $308 Million to Keep the Lights On in a Building From 1945.

  • Federal Programs Help People With Disabilities Find Work. The Handoff Can Take Months.

    The federal government has programs designed to help people with intellectual and developmental disabilities find and maintain jobs.

    It also has procedures for deciding which program is supposed to pay for that help.

    Sometimes the second part gets in the way of the first.

    A new Government Accountability Office report examining disability employment services in Georgia, Pennsylvania and Washington found that providers encountered administrative confusion, reimbursement delays and gaps in services as individuals moved between state Vocational Rehabilitation programs and employment supports financed through Medicaid.

    In one particularly striking example, a Washington provider association told GAO that funding authorizations for job-placement contracts that should take a week or less could instead take six to nine months because of staffing shortages.

    For someone who is ready to work, that is a rather long administrative definition of “ready.”

    Two Programs, One Worker

    Vocational Rehabilitation, or VR, is a joint federal-state program overseen nationally by the Department of Education’s Rehabilitation Services Administration. Medicaid Home- and Community-Based Services, or HCBS, is overseen federally by the Centers for Medicare & Medicaid Services.

    Together, they represent the primary federal programs funding employment services for people with intellectual or developmental disabilities, according to officials interviewed by GAO.

    The scale is substantial.

    Federal VR grants to states totaled approximately $3.7 billion in fiscal year 2023, and people with intellectual or developmental disabilities represented about 22 percent of VR cases nationally that year. Separately, a federally funded study estimated roughly $1.5 billion was spent on employment services for people with I/DD in FY2023, with Medicaid HCBS supplying most of that money. That estimate did not include the VR grants.

    The programs perform different roles, and on paper the transition between them is fairly orderly.

    VR typically finances shorter-term services such as job discovery and placement. Once an individual reaches job stabilization, generally after 90 days of employment, the VR case closes and Medicaid HCBS can finance longer-term assistance such as job coaching.

    If the worker loses or changes jobs, however, the process can begin again.

    That handoff is where things can become complicated.

    Mind the Gap

    GAO interviewed 12 employment service providers across the three states. Six reported service gaps caused by administrative burden or confusion between VR and state I/DD agencies.

    In Pennsylvania, one provider told investigators that support coordinators sometimes waited for a formal VR closure letter before beginning Medicaid-funded job coaching, even though individuals became automatically eligible after reaching 90 days of job stability.

    GAO said the practice caused “unnecessary gaps in services.”

    Providers can also find themselves financing the gap.

    Four of the 12 providers told GAO they experienced reimbursement delays as individuals transitioned from VR to I/DD services. One Georgia provider reported continuing to serve individuals without reimbursement during a three-month transition period while the agencies completed closure letters and prior authorizations.

    According to GAO, providers are not reimbursed for services delivered while an individual is caught between agencies after the VR case has closed.

    The consequences do not necessarily remain confined to accounting departments.

    One Washington provider told GAO that some families simply give up on obtaining employment services because of the administrative workload and back-and-forth between agencies. A Georgia provider described a woman who lost her job and then experienced months of delays reopening her VR case because of agency miscommunication.

    Pick Your Payment Problem

    Behind the coordination problem is another peculiarity: the two systems generally pay providers differently.

    VR agencies in the three states generally used a milestone model, paying providers when an individual achieved specified employment outcomes. Medicaid HCBS programs primarily used fee-for-service, paying according to units of service or time.

    Each approach comes with its own incentive problem.

    Under milestone payments, providers can perform substantial work without receiving full reimbursement if the worker fails to reach the prescribed outcome, even when circumstances are outside the provider’s control.

    GAO gives a particularly revealing example. One provider supported an individual whose workplace went out of business before the worker could demonstrate job stability. Because the required milestone wasn’t reached, the provider faced the reimbursement risk.

    Pennsylvania officials told GAO that providers considered milestone payments too risky compared with fee-for-service arrangements.

    Fee-for-service solves that problem and introduces another.

    The more services a provider delivers, the more it gets paid.

    Yet the objective of supported employment is generally for workers to become increasingly independent, with on-the-job assistance fading over time. GAO notes that fee-for-service therefore does not financially incentivize providers to reduce or eliminate services.

    In other words, one model can penalize providers when a desired outcome isn’t achieved. The other can reward them for continuing to provide the very services the program hopes workers will eventually need less of.

    Federal program design has apparently discovered diversification.

    Washington Has Been Told Before

    None of the coordination problem is entirely new to the federal government.

    In 2022, the Labor Department’s Office of Disability Employment Policy, Education’s Rehabilitation Services Administration, CMS and other federal agencies jointly encouraged state and local governments to coordinate funding through techniques known as blending, braiding and sequencing.

    The guidance also encouraged interagency agreements specifically to prevent confusion over responsibilities and delays in services.

    Georgia and Pennsylvania officials told GAO they found the guidance useful.

    Washington officials told GAO they had not yet implemented it.

    States have made other changes. Pennsylvania, for example, established guidance allowing an individual to proceed to Medicaid HCBS-funded employment services if VR has not made an eligibility determination within 120 days. Georgia officials said their VR and I/DD agencies now meet weekly to track transitioning cases and monthly to troubleshoot stalled ones.

    Those efforts matter, because GAO’s report is not a nationwide census of disability employment programs.

    The auditors examined only three states and interviewed 12 providers. GAO explicitly cautions that the results cannot be generalized to all states or all service providers.

    But within that limited sample, the pattern is difficult to miss.

    The government has programs to help people with intellectual and developmental disabilities enter and remain in the workforce. It has federal money supporting those programs. It has guidance explaining how different funding streams can be coordinated. And states have procedures intended to move people from one system to another.

    Yet GAO still found workers encountering service gaps, providers waiting months for reimbursement and families navigating enough administrative machinery that some reportedly abandoned the process altogether.

    For a system designed to remove barriers to employment, the paperwork occasionally appears to have erected a few of its own.

    Source: U.S. Government Accountability Office, Disability Employment Services: Factors Affecting Coordination Between Vocational Rehabilitation and Medicaid Programs (GAO-26-106499, September 2026).

    Fediverse reactions

  • Federal Register Watch: September 29, 2026

    The Federal Register’s September 29 issue ran to 116 documents, and the one that will matter most in a year is a pair of Federal Reserve proposals that start building the rulebook for bank-issued stablecoins. The same issue also carried two final rules, from the Education Department and the Small Business Administration, that took effect the same day they were published and skipped public comment entirely, plus a thirty-year-old advisory committee that needed a full “reestablishment” instead of a routine renewal.

    • Federal Reserve: two proposed rules start building the GENIUS Act stablecoin regime. The Fed proposed a general framework for supervising the “permitted payment stablecoin issuers” that fall under its jurisdiction, including a ban on bundling stablecoin services with other bank products, and a companion rule spelling out exactly how a state member bank asks for approval to let a subsidiary issue a stablecoin. Congress passed the GENIUS Act to bring stablecoins under bank-style oversight for the first time, so these are the rules that will actually decide which banks can get into the business and what strings come attached; because this is the first real test of a brand-new statute, how the Fed defines its terms here will shape the stablecoin market for years. (Implementing the Federal Reserve Board’s Responsibilities Under the GENIUS Act, document 2026-19860; Application Procedures for Board-Supervised Insured Depository Institutions Seeking Approval for a Subsidiary To Issue Payment Stablecoins, document 2026-19899)
    • Education Department repeals the 2024 Title IX rule, effective the same day, with no public comment. The Department’s final rule undoes the prior administration’s 2024 Title IX regulations and restores the text that existed before that rulemaking, touching more than twenty sections of 34 CFR Part 106. It took effect on September 29, the day it was published, because the Department invoked a “good cause” exception, arguing that notice-and-comment would be “contrary to the public interest” since it would delay updating the CFR. Comment periods exist so schools, students, and advocacy groups can flag problems before a rule takes hold; skipping one entirely on a rule this consequential, and making it effective immediately rather than the usual 30 or 60 days out, is the kind of shortcut that tends to draw a legal challenge over whether “good cause” really applied. (Recodification of Title IX Rules, document 2026-19929)
    • SBA strips affirmative-action provisions from its nondiscrimination rules, also effective immediately with no comment period. The rule rescinds or revises four provisions of 13 CFR Part 112, arguing the existing language encourages racial classifications that aren’t “narrowly tailored” enough to survive current equal-protection case law. SBA used a different legal shortcut than Education did, an Administrative Procedure Act exception for rules touching federal loans, grants, and benefits, but landed on the same result: no advance notice, no chance for small businesses or grant recipients to comment on a rule that changes the legal standards they have to follow, and it applies as soon as it’s published. Two same-day-effective, comment-free final rules landing in the same day’s Register, on unrelated topics but the same procedural pattern, is worth watching as a signal of how this administration is choosing to move fast on rules it considers legally exposed. (Removing Constitutional Concerns From SBA Programs, document 2026-19878)
    • Interior “reestablishes” a 30-year-old Alaska oil-spill advisory committee instead of simply renewing its charter. The Exxon Valdez Oil Spill Public Advisory Committee, created by the 1991 legal settlement between the U.S. and Alaska, has advised on commercial fishing, tourism, subsistence hunting, and wildlife recovery in Prince William Sound continuously since 1992. Federal advisory committees are normally kept alive through routine rechartering under the Federal Advisory Committee Act; needing a full “reestablishment” for a body with that long and continuous a track record suggests its charter lapsed rather than being renewed on schedule, though the notice itself never explains why. Public comments on the reestablishment are due October 6, about a week after publication. (Reestablishment of the Exxon Valdez Oil Spill Public Advisory Committee, document 2026-19933)
    • Justice Department raises immigration court fees under this year’s budget law. The Executive Office for Immigration Review announced the fiscal year 2027 inflation adjustment to fees immigrants and their attorneys pay on certain DHS forms filed in immigration court, fees that exist because of a provision in this year’s budget reconciliation law rather than a standard appropriation. Because these fees now fund part of immigration court operations directly, the size of each year’s adjustment is a small but real gauge of how much the administration is leaning on user fees, instead of congressional appropriations, to keep the immigration court system running. (Inflation Adjustment for EOIR OBBBA Fees for Certain DHS Forms; Fiscal Year 2027, document 2026-19841)

    This edition reflects all 116 documents published in the Federal Register on September 29, 2026. Excluded as routine: standard OMB/PRA information-collection notices (roughly two dozen), SEC self-regulatory-organization rule filings (eleven, from exchanges and clearinghouses including Nasdaq, Cboe, IEX, LCH SA, and the Depository Trust Company), antidumping and countervailing-duty proceedings (four, covering methionine, oil country tubular goods, wooden fence pickets, and boltless steel shelving), routine safety-zone, fishery, and environmental-review notices, Sunshine Act meeting announcements, individual DEA registration Decision and Order notices, Privacy Act computer-matching notices for state Lifeline eligibility verification, disaster-declaration notices, and other individually routine administrative filings.

  • New Mexico’s $74 Million Rural Health Announcement Is the First Piece of a Much Bigger Experiment

    The Trump administration announced $74 million Sunday to expand rural healthcare across New Mexico. That number is real, but it is also roughly one-third of the story. The full first-year award to New Mexico under the new federal Rural Health Transformation Program is approximately $211.5 million. And in Year One, a significant portion of that money is not going toward new doctors, new clinics, or new hospital beds. It is going toward building the regional architecture that will eventually decide where those things should go.

    What the $74 Million Actually Buys

    The Centers for Medicare & Medicaid Services announced that the $74 million will fund six Regional Hub Organizations under Healthy Horizons, one component of New Mexico’s broader rural health strategy. The hubs are charged with identifying local needs, establishing priorities, and developing plans for expanding specialty, maternal, behavioral, and chronic-disease care. Year One, in other words, is substantially a planning operation — which may be exactly what New Mexico needs.

    New Mexico’s own Legislative Finance Committee has already raised that question in a different context. A committee analysis found that lawmakers had appropriated $226 million to the state’s Rural Health Care Delivery Fund between 2023 and the 2025 special session, and that by October 2025 slightly more than a quarter of it had been spent. Legislative analysts found no overarching plan guiding those investments and questioned whether grant decisions were adequately tied to demonstrated community need. Healthy Horizons is, at least structurally, an attempt to build that plan before spending the rest of the money.

    The Funding Stack

    The Rural Health Transformation Program was created by the 2025 federal budget reconciliation law, Public Law 119-21, with $50 billion appropriated over five fiscal years — $10 billion annually from FY2026 through FY2030. Half of each year’s funding is split equally among states with approved applications; the remainder is distributed based on rural population, healthcare infrastructure, state policy, and program design. All 50 states received awards. New Mexico’s first-year share is $211.5 million, of which $74 million covers the Healthy Horizons hub phase. The program’s full five-year budget for New Mexico, per CMS’s published application abstract, is approximately $393.3 million — covering workforce recruitment, telehealth, remote patient monitoring, health-data infrastructure, and chronic-disease management alongside the specialty and maternal care components.

    The Larger Context

    The same law that created the Rural Health Transformation Program also made substantial changes to Medicaid financing. New Mexico officials and members of its congressional delegation have warned those provisions could place additional financial pressure on providers serving Medicaid patients. The rural health money and the Medicaid changes are separate components of the same legislation — the $74 million is not replacement funding for reduced Medicaid flows — but the juxtaposition is unavoidable. Washington is restructuring one major source of rural healthcare revenue while simultaneously running a $50 billion experiment to transform how rural healthcare is delivered. CMS has a dedicated Office of Rural Health Transformation to administer the program through 2031 and has emphasized both state flexibility and accountability for results.

    New Mexico has spent hundreds of millions of dollars on rural health in recent years, and its own legislative analysts subsequently found the investments lacked strategic direction. Six regional hubs are now being stood up with federal money to produce exactly that direction. Whether they do — and whether the infrastructure being built in Year One translates into measurable outcomes by Year Five — is the question the announcement doesn’t answer. It just starts the clock.

    Sources: CMS, Rural Health Transformation Program — New Mexico Application Abstract; New Mexico Legislative Finance Committee, FY2027 Budget Recommendations.

    Fediverse reactions

  • The Reform Congress Wouldn’t Pay For

    The federal government has spent a decade trying to prove it can sell its own buildings faster than it can grow moss on them. In 2016, Congress passed the Federal Assets Sale and Transfer Act, a law with a genuinely tidy premise. Create an independent board to identify unneeded federal real estate, let the General Services Administration sell it, and use the proceeds to fund the next round of sales. A self-perpetuating disposal machine, powered by its own output.

    Then Congress built the machine, filled the tank once, and walked away for three years.

    A Fund in Name Only

    FASTA created the Asset Proceeds and Space Management Fund, a pool meant to hold sales revenue and reinvest it in future disposal costs. It is, on paper, an elegant piece of institutional design. Sell an unneeded federal warehouse, use a slice of the proceeds to relocate the tenants of the next unneeded building, repeat until the government’s real estate footprint stops costing taxpayers billions a year in upkeep on buildings nobody uses. A recent GAO report found that the government owns hundreds of thousands of buildings, and that disposing of the ones nobody needs can take years or decades while the bills keep coming.

    The catch, tucked into every appropriations-dependent fund since the invention of the federal budget, is that money sitting in an account is not the same as money an agency is allowed to spend. Congress has to appropriate it first, round by round, year by year, at its own discretion. FASTA’s authors either didn’t anticipate how easily that discretion could evaporate, or didn’t care.

    The Three Years Nobody Noticed

    Between 2016 and 2022, Congress appropriated a cumulative $90 million from the fund, a modest but functioning start. Then, from fiscal year 2023 through 2025, it appropriated nothing. Zero dollars, three consecutive years, while GSA’s High-Value Round sales kept depositing proceeds into an account it increasingly could not touch. By the end of fiscal year 2022 the fund held roughly $194 million in sales revenue. The government had money sitting in the bank for a program built entirely around the premise of using its own proceeds efficiently, and it declined to authorize a withdrawal.

    It is worth sitting with that for a moment. This isn’t a story about an agency losing track of paperwork or missing a deadline. It’s a story about a fund functioning exactly as designed, generating exactly the revenue it was supposed to generate, and then being denied access to that revenue by the same legislative body that wrote the design. The fund didn’t fail. It was throttled.

    In 2026, with the program’s last recommended round underway and the Public Buildings Reform Board set to dissolve at year’s end, Congress appropriated roughly $143 million more, bringing the cumulative total to $233 million against a fund that, by then, held about $538 million following the long-delayed sale of a California property called Laguna Niguel. The gap between what the fund held and what GSA was permitted to spend ran as high as $300 million at points between 2022 and 2026. That gap is not an accounting curiosity. It is nine months of stalled relocation for the Railroad Retirement Board, an agency officials said would find self-funding “devastating to its mission.” It is a Chicago federal building whose tenant is waiting on a $19.5 million allocation that GSA has repeatedly flagged as “in flux.”

    Self-Funding Was Always the Fine Print

    GSA didn’t sit still during the drought. The agency requested funding for a separate initiative called the Optimization Program, eventually lining up roughly $470 million across fiscal years 2025 and 2026 to plug the gap FASTA’s own fund wasn’t allowed to fill. But that money comes with its own leash. Projects above a statutory cost threshold require congressional prospectus approval before the funds can even be obligated, and as of August 2026, only the House side of that approval had come through. The Senate side hadn’t. So GSA now sits on two separate pots of money it can see but, in one case, still cannot spend, for two separate bureaucratic reasons layered on top of each other like sediment.

    Meanwhile, the law itself quietly changed the plumbing. Amendments passed in 2025 now route proceeds from the program’s second round and beyond directly to whichever agency owned the building, bypassing the central fund entirely, still subject to the same appropriations process that starved the fund in the first place. GSA officials told auditors the change granted no new efficiencies. It just moved the choke point somewhere else.

    The Self-Funding Machine, Unfunded

    None of this required a scandal, a whistleblower, or a single act of malice. It required Congress to write a law premised on reinvested proceeds, then treat the reinvestment step as optional for three budget cycles running. The Public Buildings Reform Board will cease to exist by the end of this year, having recommended more disposals than the appropriations process was willing to underwrite. What’s left is a machine that works exactly as designed, sitting mostly idle, waiting on a permission slip that took Congress three years to sign.

    Source: U.S. Government Accountability Office, Federal Real Property: GSA Has Made Progress Disposing of Unneeded Properties, but Challenges Remain (GAO-26-108451, September 2026).

    Fediverse reactions