SI Agent News
RumorPolicySecurityBusinessResearch

US AI Safety Center staff asked to report informally to White House

The move toward closer ties between the Commerce Department's AI evaluation center and the White House has sparked concerns over the independence of federal safety research.

$10 millionannual budget for the center

The short version

Why it matters

This development raises questions about whether federal AI safety evaluations will remain objective or become subject to political influence. For researchers and companies relying on government standards, the shift could impact the transparency and reliability of future safety benchmarks.

A shift in reporting lines

Staff at the Center for Advancing Innovation and Standards for Super Intelligence (CAISSI) were recently asked to report informally to an office within the White House. This information comes from two former government officials who spoke anonymously to the Associated Press. The report does not describe a formal reorganization of the agency's structure.

CAISSI is currently an independent division of the Commerce Department. It was previously known as the U.S. Center for AI Standards and Innovation (CAISI). The move to an informal reporting structure with the White House has not been confirmed by a formal directive or official notice from the Commerce Department.

The White House did not respond to specific questions regarding the center. Instead, the administration pointed to the President's creation of a "Super Intelligence Force" and a report commissioned from the Council of Advisors on Science and Technology. This lack of specific detail leaves the exact nature of the reporting obligations unclear.

Leadership and publication gaps

The center has faced significant leadership instability over the past year. A director was hired in April but was fired within days of starting the role. The individual who replaced that director resigned after only a few months, leaving the center without a permanent head.

Currently, the head of the parent institute is serving as the acting director. This lack of permanent leadership coincides with a notable change in the center's output. The center has stopped publishing its research, a move that has drawn attention from lawmakers.

In August, members of Congress began asking for explanations regarding why the center stopped making its reports public. This pause in transparency is a key point of concern for those who rely on the center's work to establish safety standards for super intelligence.

The debate over independence

The move toward closer ties with the White House has created a divide in opinion among observers. Some critics express concern that these closer connections could lead to the politicization of AI safety decisions. They worry that findings might be muted or altered to align with specific political goals rather than technical realities.

On the other hand, some observers argue that proximity to the White House could be beneficial for the center's mission. They suggest that a closer relationship with the executive branch could provide the division with greater prominence. This increased visibility might also lead to more significant funding and resources for AI safety research.

Because the reporting arrangement is informal, it remains unclear if any specific findings have already been changed. The lack of a formal reporting line means that the center's independence is currently a governance question. The situation highlights the tension between needing executive support and maintaining scientific integrity.

Impact on the research community

For researchers and research offices, these changes could affect how they interpret government evaluations. Teams that rely on government pre-deployment testing or published evaluation methods should be aware of the center's current status. Since the center has paused public reports, the availability of new data is limited.

Organizations that cite federal evaluations in their safety cases or grant proposals may need to be more diligent. They might want to record exactly which body produced a specific result and when it was released. This is especially important as the center's reporting structure becomes less clear to the public.

The broader AI safety field will likely watch for formal documents to clarify the situation. A written directive from the Commerce Department or the White House would provide clarity. Additionally, the appointment of a permanent director or a resumption of public reports would settle many of the current questions.

What to watch for next

The most important next step is the publication of any formal documents regarding the center's structure. A written notice from the Commerce Department would clarify whether this is a permanent shift in how CAISSI operates. Until then, the arrangement remains an informal reporting line based on anonymous accounts.

Observers should also keep an eye on the center's research output. Any resumption of public reports would be a significant signal for the research community. It would indicate whether the center is moving back toward its original role as a public-facing standard-setter.

Finally, the appointment of a permanent director will be a key milestone. A stable leader could provide the consistency needed to navigate the center's relationship with the White House. This leadership will determine how the $10 million annual budget is spent and which safety priorities are pursued.

What to watch

Try it

Look up the previous name of the center, CAISI, to see what types of pre-deployment testing they were known for before they paused their reports.

Sources and links

Liked it? Say why

Add a line for other readers.

No account needed. We never show your email. Comments with links, ads or abuse are held and not shown. Your comment shows after a quick check.

Listen to the AI Wire headlines

Listen to the AI Wire

Listen to our daily AI shows every morning: on your commute, in the shower, wherever you are

Listen to the daily shows

First reported by CASRAI · Read the original. Written by the SI Agent News Desk from the reporting above.Written with AI from the source above. Every number is checked against the source. Spot a mistake? Corrections.

More AI news