Pinned Tweet
Was great to talk with @AU_SIS about how my experience was shaped by my education with AU, and looking back at my service and how I hope I’m still giving back.
Air Force veteran @JasonBakerJB, SIS/MA ’19, spent 20 years carrying out U.S. foreign policy. At SIS, he brought a practitioner’s eye to the classroom. Now he’s working on one of today’s fastest-moving national security challenges: AI. american.edu/sis/news/sis-ve…
10
925
Jason Baker retweeted
The White House just held a background briefing to address the string of tech outages and unanswered questions about what's going on. The senior administration official said:
— What has happened is not the result of cyber attacks, but due to the “hardening and upgrading of our systems," including “systems being patched, hardware being applied.”
— The official said: “It’s good the American people are focused on the need to upgrade our legacy systems.”
— Asked about the prospect that if this is what happens when upgrades made, what should people expect if there is an actual cyber attack, the official said: “That is a question the American people should be asking.”
Jason Baker retweeted
Here's what @andrewrsorkin said on stage at @dealbook about VP JD Vance abruptly pulling out of the NYC conference and staying in DC today:
“We were going to be joined by the Vice President, but I am sad to report that we got a call literally probably now 20 or 30 minutes ago that he is not going to be able to join us today. He was called to stay back in Washington. He was supposed to be wheels up this morning. As you know, Secret Service, because you lived through the lines, has been here all night, readying for his entrance here at Dealbook. All of this comes, of course, amid a series of outages from Microsoft, the NSA, the airline ground stop, JPMorgan, and Zelle, and I am told that he is needed in Washington to deal with a series of meetings that are taking place as we speak.”
5 days left to apply for this amazing opportunity to pivot into AI policy! Education, research support, and more networking than you can shake a stick at!
Apply for GovAI's 2027 DC Summer Fellowship!
📍 Location: Washington, DC
📅 Deadline: Sunday, November 22nd, 2026 at 11:59 pm ET
The DC Summer Fellowship is a three-month bipartisan opportunity designed to accelerate or launch impactful careers in American AI governance and policy. Participants deepen their understanding of the field, connect with a network of experts, and build their skills and professional profile, all while conducting a research project of their choice under expert guidance.
🔗 Learn more: governance.ai/post/dc-summer…
Jason Baker retweeted
I understand why Americans are concerned about the widespread outages we’ve seen in recent days, affecting everything from banks and major technology companies to critical government systems. At this point, there is no indication that these incidents are the result of a cyberattack, foreign interference, or some coordinated effort.
What appears to be happening instead is a consequence of how quickly artificial intelligence is changing the cybersecurity landscape.
AI is making it possible to identify vulnerabilities across enormous numbers of systems at unprecedented speed, creating enormous pressure on companies and government agencies to patch those vulnerabilities immediately, often all at once, putting tremendous strain on systems that weren’t built to handle that pace of change.
While we still need to get to the bottom of these incidents, the evidence so far points toward legitimate efforts to secure our systems, not malicious activity.
Still, these disruptions should serve as a wake-up call. It feels like every day brings another reminder that AI is moving faster than our ability to manage its consequences.
This is exactly why I’ve been pushing for commonsense guardrails on AI and stronger cybersecurity protections.
Unfortunately, President Trump has shown little interest in confronting these challenges, his administration has repeatedly undermined its own credibility, and its cuts to CISA have weakened the very expertise we need to navigate moments like this.
Americans deserve clear, reliable information and a government that is prepared to respond. I’m going to keep digging into these incidents, pressing for answers, and making sure we’re doing everything possible to protect our critical systems. But based on what we know today, there is good reason to believe these disruptions are the unintended consequences of efforts to make our systems safer, not evidence of a broader attack.
Jason Baker retweeted
UPDATE: The NSA has issued a statement
"The National Security Agency experienced temporary degradation during scheduled routine maintenance. While some internal systems were affected, mission-critical capabilities remain fully operational and our ability to defend national security continues. NSA's infrastructure is designed with multiple layers of redundancy to maintain operations during planned maintenance or unexpected events."
We break: The highly classified network that the National Security Agency uses for its internal communications is down, according to five people familiar with the matter.
The network, called NSA Net, allows employees to communicate classified information internally. At this point there is no evidence of a cyberattack, sources said.
As the agency works to understand the scope of the problem, it has enacted staggered arrivals for personnel, the sources said.
CNN has asked the National Security Agency for comment.
cnn.com/2026/11/17/politics/…
Jason Baker retweeted
There we go. “A person familiar with the incident said the NSA was simply updating its software to patch the system, adding that NSANet would be offline only temporarily and that there was no evidence of a cyberattack.” (wsj.com/politics/national-se…)
So both the FAA incident and this one were both down to upgrades/updates gone wrong, rather than malicious action or AI run amok.
I still believe the majority of this stuff is unconnected to either hostile cyber activity or to AI, but there is a slightly end-times feel to stuff at the moment.
"The highly classified network that the National Security Agency uses for its internal communications is down, according to six people familiar with the matter. The network, called NSA Net, allows employees to communicate classified information internally. At this point there is no evidence of a cyberattack, sources said."
cnn.com/2026/11/17/politics/…
I think this is quite possibly the best explanation right now.
There are lots of outages... Why? AI attacks?
My more normal guess is that because there have been a lot of AI-discovered cyber vulnerabilities (including lots probably not yet disclosed), companies across all industries are having to do a lot of patches and upgrades, probably faster than usual. Such upgrades can sometimes go wrong, causing outages.
2
13
2,955
Jason Baker retweeted
Good news from the last few incidents over the last several days, everyone in public statement seems very sure that:
“at the time of writing, is not down due to a cyber attack.”
It’s all just the spotlight effect… no reason to expect otherwise…
I keep repeating this to myself.
Hopefully hyperstitioning it into existence!
According to a report from CNN, citing people familiar with the matter, the National Security Agency’s (NSA) secure internal communications network, NSA Net, is down. Per the report, the network, which allows secure classified communications internally, at the time of writing, is not down due to a cyber attack. The NSA has also reportedly implemented a tiered arrival framework for employees today.
And now a reading from the book of Nathan, chapter 1:
"And if thou dost lay eyes upon two ants in thine kitchen, is it not true the full measure of ants is far greater than two?"
A bunch of crucial systems going down within the span of a few days (FAA systems leading to third ever nationwide ground stop, several large banks, internal NSA comms) with all of them saying no evidence of a cyberattack?
16
922
Breaking: Everything, apparently.
Breaking: The NSA's classified internal comms network—NSANet—is down, sources tell @NatashaBertrand @ZcohenCNN @snlyngaas & me. Disruption is significant enough the agency has staggered arrivals. No indication right now of a cyberattack but officials still working to understand.
1
184
Jason Baker retweeted
The Trump administration's new rule limiting China's access to America's computing power is an important step toward securing America's AI lead. Congress must now build on that progress by passing the Remote Access Security Act to close the cloud loophole. bloomberg.com/news/articles/…
Jason Baker retweeted
Today in the @washingtonpost, I argue that divided government can actually create opportunities for a balanced policy on AI, and I outline three areas for cooperation that are essential to continued US leadership: energy, high-skilled immigration, and governance.
Jason Baker retweeted
DHS Secretary Mullin says he finds its plausible after receiving relevant briefings that this recent spate of tech outages is downstream of integrating AI/SI systems that overwhelm older platforms
MULLIN ON OUTAGES: I talked to one guy in the tech industry, he asked not to mention his name, it's not Elon. I asked, what could be causing these outages? He said I believe it's SI, because these programs aren't designed to work as fast as SI is.
So when they put in a program that's able to write codes and it goes in and self-fixes, as it starts to self-fix itself it like a change order on a house.
That kind of makes sense to me.
Jason Baker retweeted
Very, very strange statement tonight from @CommerceGov and @TheJusticeDept about Megaspeed. The facts here just don't add up, and this does nothing to explain why Commerce did not cut Megaspeed off from chips. In fact, it raises even more questions about what happened here.
BIS claims that it started an investigation into Megaspeed in 2024, and following an end-use check in June 2025 it handed off the criminal investigation to DOJ. It also said that in June 2025, Nvidia put Megaspeed on its own internal no-trade blacklist--implying this solved the issue and cut them off from Nvidia chips.
The problem with this story, is that Megaspeed kept buying chips after Nvidia put them on their no-trade list. @mackhawk reported in December 2025 that Megaspeed imported 17,544 GB300 AI chips in November 2025, worth about $787 million. There are multiple other reports that say Megaspeed continued to receive chips into 2026.
If Commerce was so concerned about Megaspeed smuggling chips to China that the DOJ opened a criminal case after an end-use check in June 2025, and Nvidia also put them on an internal blacklist, why didn't Commerce also put them on the Entity List? That is literally the entire purpose of the Entity List, to list companies in situations like this! If it had done that--and also put Megaspeed's affiliates and subsidiaries such as Speedmatrix (its Malaysian subsidiary) and PT Megaspeed International (its Indonesian subsidiary) on the Entity List--Commerce could have actually cut Megaspeed off from chips!
Instead, Commerce appears to have outsourced Entity List restrictions and export enforcement to Nvidia. That is insanely irresponsible! Does @USTreasury outsource sanctions enforcement entirely to US banks? Of course not, OFAC sanctions entities that are violating US sanctions. How is "we didn't act because Nvidia acted for us" the best defense Commerce can muster here?
And also, why was Nvidia's blacklisting action so ineffective? Did Nvidia also place Megaspeed's subsidiaries such as Speedmatrix and PT Megaspeed International on its blacklist? I would bet they did not! If they kept shipping to them after blacklisting Megaspeed, that would be very concerning. What is the point of Nvidia blacklisting Megaspeed if they continue to allow their chips to be shipped to Megaspeed's subsidiaries in Malaysia and Indonesia?
It's hard to decide who screwed up more here, @CommerceGov or Nvidia. Just a total failure all around. How could both sides have agreed that Megaspeed represented a national security risk, but completely failed to prevent Megaspeed from buying $787 million worth of chips just five months after it was blacklisted? And undoubtedly more chips in 2026.
Maybe instead of writing defensive press releases that don't add up, @CommerceGov could write a rule finally placing Megaspeed on the Entity List. But maybe they're waiting for Treasury and NSC to write that rule for them as well.
Jason Baker retweeted
Great to be on @CNN with @jimsciutto tonight discussing the impact of Commerce not enforcing export controls: we would probably be a year further ahead of China in AI if they couldn’t access US AI chips. The cost of Commerce refusing to close loopholes until today was huge.
!!: Question: Is there a way to quantify the cost of Nvidia selling advanced chips to companies tied to China?
CFR’s @ChrisRMcGuire: “It’s probably cost us a year of advantage in the AI race.”
Jason Baker retweeted
We’re entering a new and dangerous era of artificial intelligence.
We need stronger tools to protect against consequences that could be both profound and irreversible.
That means responsible regulations that protect ratepayers and ensure every American benefits from this technology, not just the companies creating it.
Jason Baker retweeted
Some initial takes after reading this announcement from OpenAI on new embedded assessments:
These seem like reasonable collaborations and I'm glad OpenAI is doing them. All four of these orgs care deeply about AI safety and have talented researchers, and aren't just corporate rubber stamps. But none of the collaborations seems to clearly involve "employee like access," (desks, badges, laptops — not set by the minimum info the evaluations need) to prevent catastrophic risks as Sam promised in September.
For instance, the details of the collaboration with METR seem extremely narrow, only focusing on evaluating misaligned behavior from a large RL run. OpenAI also say that they have "preliminarily agreed to conduct another monitor assessment which we intend to be broader, but this is pending mutual agreement on the assessment scope"
A footnote on the post also says more explicitly regarding the collaboration with METR on their RL run:
"This scope is less expansive than what METR outlines in their alignment assessment proposal, which envisions a broad survey of misalignment."
From this section it seems quite clear that this is not an announcement to give METR employee level access - the other collaborations are fewer on details but also appear sufficiently narrow in scope that I don't expect them to have much of an effect at identifying areas where OpenAI's current safety and security practices are insufficient or dangerous.
Apollo made a similar point in a tweet after the announcement today:
“This pilot is an ambitious first step towards the goal of obtaining those strong guarantees, but its access and disclosure rights don't yet meet our bar for full embedded evaluations.”
It's possible that this announcement literally fulfills the letter of Sam's prior commitment (insofar as e.g. OpenAI is basically hiring Goodfire staff to do interpretability research internal to OpenAI), but it does not at all meet the spirit of having independent researchers with broad access able to identify and address imminent risks to the public. In Dario's underlying post that Sam favorably commented on and offered his commitment, Dario describes the role of embedded evaluators "whose role is to verify adherence to safety practices and commitments, report incidents, and help assess the alignment of not just completed AI models but training pipelines and processes." It is extremely unclear to what extent any of these evaluators will be performing those functions. (To be clear, i’m not sure that Anthropic is fulfilling these commitments either, at least in their announcements to date and the draft legislation they have put forward.)
This only more clearly shows the need for these practices to be mandated, instead of entirely voluntary and on the company's terms (and remember of course many of OpenAI and Anthropic's competitors are just not doing anything at all in terms of strong independent evaluators with meaningful access). Even as I was writing up this post critiquing this announcement, I was very aware of the fact that OpenAI people may reasonably react “what is the point of doing this at all” if people are going to criticize whatever we do as not enough and orgs doing nothing get a free pass. So let me be very clear that OpenAI is doing much much more here than Google, xAI, Meta, or any of the Chinese developers (some aspects of this are better than what Anthropic has done, some - like the amount of access given to METR - is worse).
I also realize that getting even these limited (but extremely valuable) collaborations across the line took an immense amount of work internally and i'm grateful for everyone who worked to make them happen to give, both within and outside the companies. But I also think its worth being very clear about areas where this still falls short, both from OpenAI's own commitment and above all what is objectively needed for sufficient transparency and safety.
We’re working with @ApolloResearch, @METR_Evals, @TransluceAI, and @GoodfireAI on embedded assessments: giving outside experts deep access to our internal models and systems to investigate alignment and test our safeguards.
These assessments build on our ongoing alignment and safety research, bringing independent scrutiny to how we identify and manage risks in our systems.
openai.com/index/our-first-p…
Jason Baker retweeted
More evidence on how much Chinese AI startups rely on distillation: HSBC just cut Zhipu's price by 2/3rds because of "increased competition, rising R&D costs, and the strengthening of anti-distillation measures by leading US AI labs."
Awesome summary and discussion by Tim and I agree: It's pretty good!
The Commerce Department has released a major new export control rule targeting China’s access to advanced AI compute.
It’s pretty good! Here’s what it does:
(1) Restricts Chinese access to overseas AI clusters (seeks to address one of China’s largest and fastest-growing sources of AI compute)
This is a new worldwide end-use control, which we’ve previously recommended (third option in image below).
A license is required if providing cloud AI compute to Chinese or D:5-headquartered countries, or if acting as an intermediary training models for these actors.
Technical requirements are:
- Applies to any data center or group of “linked data centers” or providing more than 12 million aggregate TPP of compute (the equivalent of around 760 H100s) of AI chips meeting the current controlled thresholds.
- “Linked data centers” are defined as sharing physical or virtual computing resources, jointly balancing workloads, or replicating data as a unified whole. So this is intended to capture multi-data center training runs.
This applies everywhere except the US. It also covers existing data centers, which have a 90-day deadline to get a license.
(2) New controls for AI chips going to Malaysia, Singapore, Indonesia, Thailand, and Brazil (seeks to address smuggling of AI chips to China via third countries)
The evidence for large-scale smuggling of AI chips to China via Malaysia and Singapore is strong, and there’s also evidence this is happening via Indonesia and Thailand. Brazil, on the other hand, is a bit of a head-scratcher.
This is a case-by-case license review policy, rather than a presumption of denial (that's the policy for chips going directly to China/D:5 countries).
BIS will evaluate each license on national security grounds.
There’s an exclusion for qualifying US-headquartered recipients, subject to reporting requirements.
(3) Formally repeals the AI Diffusion Rule
This has been a long time coming. This eliminates the country-tier system, the Validated End User authorizations (VEU) for data centers, and export controls on advanced AI model weights.
However, the replacement measures above address many of the gaps created by this, and, of course, the AI Diffusion rule wasn’t actually being enforced.
(4) Tightens controls on Chinese chip designs manufactured abroad
This modifies and strengthens the Biden-era “Foundry Due Diligence” rule, which applies to foundries and OSATs.
This seeks to prevent Chinese chip designers from circumventing export controls through overseas foundries and packaging facilities. It’s an important complement to the measures above.
—
So what are the gaps? More on this to come, but from an initial read:
(1) To determine remote-access eligibility, the rule requires due diligence on direct or indirect Chinese ownership or control. However, its licensing trigger is based around company headquarters and ultimate-parent headquarters, which could allow for some shenanigans around corporate structures.
(2) The rule doesn’t prescribe how AI compute providers must verify customers' identities, or require them to identify and verify every user accessing their compute. The rigor of these checks could vary widely across providers.
(3) The end-use restrictions don’t apply when compute is used exclusively for commercial recommender systems. So this opens a path for large-scale Chinese AI compute usage. This could be fine if verifying this kind of use is possible, but this is likely outside the current technical capabilities of many AI compute providers.
(4) The end-use restrictions don’t apply to data centers in the US. So a Chinese company can still access large amounts of AI compute via US data centers.
This is an interim final rule, meaning it would take effect upon publication in the Federal Register, while allowing 60 days for public comments.
Read the full interim final rule here: bis.gov/media/documents/rin-…
































