Breaking

Trump Says Major AI Companies Signed Voluntary Superintelligence Safety Accord

September 29, 2026 at 17:52 EDT

Policy & Regulation
Foundation Models

President Donald Trump said on Tuesday, September 29, 2026, that major AI executives had signed a voluntary accord for reviewing the development of what he calls Super Intelligence, putting industry-led controls at the center of the administration's answer to AI safety fears. The deal is voluntary and morally binding rather than statutory, with companies committing to internal and external reviews while enforcement details remain unresolved.

Sept. 29, 2026 | U.S. AI policy

Voluntary controls for Super Intelligence

Trump said major AI companies signed a morally binding accord built around self-policing, audits and board-level review. The unresolved issue is whether voluntary checks become enforceable safeguards.

Review stack

control and audit layers

Committee

~10

people discussed for oversight

Czar timing

3-4

days for a named AI czar

Portal scope

29,000+

government websites indexed

Internal

External

Board

Layer 4

4-layer stack

The audit promise is layered

Each block represents one control layer. Public descriptions clearly identify internal checks, outside audits and board review, while implementation details remain the decisive gap.

Voluntary oversight flow

Company review

Internal risk checks

→

Outside audit

External evaluators

→

Board review

Reports move upward

Innovation case

Executives framed audits and controls as compatible with fast AI development and U.S. competitiveness.

Oversight case

Critics want mandatory testing, secure evaluation environments and incident reporting for frontier systems.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.