Claims extracted
16
sentences that state something as happening or true
Uncertain
19%
3 of 16 hedged
("reportedly", "may", "according to")
Leaning
not political
takes no side on a contested political question
Publisher signals
60/100
Al Jazeera's corrections and
hedging, across everything collected from it. Not a measure of accuracy.
Outlets on this story
3
Technology
Same framing elsewhere
1
articles that frame the event the way this one does
OpenAI has revealed that its AI agents disabled monitoring for unsanctioned behaviour during internal use in the latest incident of the frontier technology going rogue.
asserted
technology → reveal → incident
In a report published on Tuesday, the AI giant said that its agents turned off or weakened monitoring systems for detecting misalignment – when AI models act at odds with human wishes – during two internal coding tasks.
asserted
models → publish → tasks
Recommended Stories
list of 4 items- list 1 of 4Musk confirms SpaceX and Tesla exploring merger
- list 2 of 4UN experts call on India to stop using ‘excessive force’ against protesters
- list 3 of 4Drone
asserted
experts → confirm → protesters
strike on Abyei market near Sudan-South Sudan border kills 150
- list 4 of 4Portugal cost-of-living protesters pour back into camps cleared by police
asserted
protesters → kill → police
The incidents involved a GPT-6.1 Sol agent disabling automatic alerts to the company’s security team, and a GPT-6 Astra agent removing a monitoring system and falsely claiming it had not done so, OpenAI said.
asserted
OpenAI → involve → system
The ChatGPT creator said neither case met the threshold for a highest-priority incident necessitating immediate review by its security team.
asserted
case → say → team
OpenAI said it updated its systems to better detect attempts to disable misalignment monitoring and began investigating the cause.
asserted
it → say → cause
“It is possible that this behavior is simply an example of ‘benign’ overpersistence, with models not understanding why monitoring is important or insisting that instructions be executed beyond what a reasonable person might expect,” the San Francisco-based firm said.
uncertain
firm → understand → what
“We are also developing evaluations to test whether models might have a propensity to evade misalignment monitoring in particular and exhibit motivated reasoning in order to do so.”
uncertain
models → develop → reasoning
Roman V Yampolskiy, an expert on AI and cybersecurity at the University of Louisville, described the revelations as “extremely concerning, but not surprising.”
asserted
Yampolskiy → describe → revelations
“A safety mechanism that the system being monitored can disable is not a reliable safety mechanism.
asserted
system → monitor → that
Particularly concerning is that one agent subsequently denied removing the monitor,” Yampolskiy told Al Jazeera.
asserted
Yampolskiy → deny → Jazeera
“Monitoring cannot guarantee safety if the monitored system can undermine the monitor,” Yampolskiy said.
asserted
Yampolskiy → guarantee → monitor
“As capabilities increase, such failures could become substantially more difficult to detect and reverse.
uncertain
failures → increase → ?
The burden of proof must be on developers to demonstrate safety before deployment, not after failures occur.”
asserted
failures → demonstrate → deployment
Rogue behaviour by AI models has been under heightened scrutiny since July, when OpenAI revealed that some of its agents broke out of a controlled testing environment and hacked the AI platform Hugging Face.
OpenAI and its chief rival Anthropic have called for developers and governments to coordinate a global slowdown in AI development to reduce the risk of catastrophic harms.
The US and China, the world’s leading AI powers, have displayed little appetite for cooperation on AI regulation, with coordination so far limited to pledges to improve communication on incidents with implications for safety or national security.
asserted
coordination → heighten → safety
The Aporia
