TRV-2026-1212Version 3 · Revised

Saved · see the current record

Why this version exists

Editorial pass: gain/problem restated as concrete outcomes.

The exact text we saved

This is the version exactly as it was saved, character for character. The SHA-256 hash below is a 64-character code worked out from that text. Run the same text through SHA-256 and you get the same code; change a single letter and the code comes out completely different. Each version also stores the previous version’s hash, which chains them together so no version can be swapped out quietly.

TRUVACE RECORD VERSION
record: TRV-2026-1212
version: 3
kind: revised
reason: Editorial pass: gain/problem restated as concrete outcomes.
timestamp: 2026-11-04T09:50:21.968573Z
status: published
lens: p_space
sector: policy
headline: Trump announces vague ‘morally binding’ AI deal among tech CEOs for ‘tremendous self-policing’
dek: Donald Trump announced on Tuesday that the heads of the largest US tech and AI companies had signed on to a “morally binding” agreement to put controls on artificial intelligence. He announced the deal, called the Joint Commitment On Frontier Responsibilities, following a White House luncheon with top tech executives. “It’s almost like a constitution in a way, and the biggest people in the world signed that, and I signed it as president, and it really ⁠is a form of protection,” Trump said. Trump also signed an e…
gain_title: (none)
problem_title: Major AI companies have inadvertently hacked outside organizations during safety tests, and the new voluntary pact leaves oversight to company-chosen auditors and boards, with no regulators or required public disclosure.
trace_subject: (none)
gain_reading: (none)
gain_evidence: (none)
problem_reading: Major AI companies have inadvertently hacked outside organizations during safety tests, and the new voluntary pact leaves oversight to company-chosen auditors and boards, with no regulators or required public disclosure.
problem_evidence: appears to carry no enforcement mechanisms or legal implications and is instead a voluntary commitment | None of the controls listed in the agreement involve government regulators or include a commitment to publicly detail the findings of independent evaluations | allow companies to pick their own evaluators, appoint their own oversight boards and decide whether to publish their results
quick_read: On September 29, 2026, President Trump announced a 'morally binding' Joint Commitment On Frontier Responsibilities signed by heads of major US tech and AI firms after a White House luncheon, alongside an executive order renaming artificial intelligence to 'superintelligence' in federal communications. The document proposes four layers of internal monitoring, internal verification, external auditing, and board review for frontier AI testing.

The development matters because it substitutes voluntary self-policing for government regulation at a time when lawmakers, researchers and some companies have called for more oversight after safety-test incidents where labs inadvertently hacked outside organizations. What remains uncertain is whether company-selected auditors and boards that may not publish findings will improve safety, given the article notes existing internal monitoring has already failed to prevent failures.
limitation: The agreement is described as voluntary with no legal enforcement, no role for government regulators, and discretion for companies to select evaluators and withhold results, while the first layer of internal monitoring already exists and has failed to prevent failures.
tag: Evidence-backed problem
key_points: Trump announced the Joint Commitment On Frontier Responsibilities after a White House luncheon with heads of largest US tech and AI companies including Elon Musk, Mark Zuckerberg, Jensen Huang and Dario Amodei. | Trump signed an executive order directing federal agencies to use 'Super Intelligence' and 'SI' and to no longer acknowledge the terms 'Artificial Intelligence' or 'AI'. | The document describes four layers: internal safety monitoring during training, an internal team to check that monitoring, an external auditor to assess controls, and an independent board to review reports. | The tech industry has faced scrutiny after incidents where major AI companies inadvertently hacked into outside organizations during safety tests.
rundown: On 2026-09-29 Trump announced the Joint Commitment On Frontier Responsibilities after a White House luncheon with top executives including Elon Musk, Mark Zuckerberg, Jensen Huang and Dario Amodei, describing it as 'almost like a constitution' and 'a form of protection'.

The same day he signed an executive order directing all departments and agencies to use 'Super Intelligence' and 'SI' in official correspondence and to no longer acknowledge 'Artificial Intelligence' or 'AI', following a similar declaration at the United Nations.

The commitment lists four layers: internal safety monitoring during training, an internal team to verify that monitoring, an external auditor to independently assess controls, and an independent board to review reports, with no requirement for government regulators or public disclosure of evaluations.
sources:
- journalism | The Guardian | https://www.theguardian.com/us-news/2026/sep/29/trump-ai-deal-tech-ceos-superintelligence | 2026-09-29
prev: 538da31c95d32f09d3b13e26f04f8adcc8643c0ec50632c97103bbf51a168171
SHA-256 hash
cf9fd4765cbc55efc5cc53de4cc86a0a88199a78f1b9501b1c0115123c0b974d
Previous version’s hash
538da31c95d32f09d3b13e26f04f8adcc8643c0ec50632c97103bbf51a168171
Check this record hasn’t been altered
How to check it yourself, without trusting this page

Download any version from /api/record/TRV-2026-1212 and hash its canonical field yourself, for example with shasum -a 256. You should get its content_hash. Each version’s text also ends with prev:and the hash of the version before it (version 1 points to 64 zeros). If anyone had changed a single character of any version after it was saved, those hashes wouldn’t line up.