Fact check
“OpenAI’s agents accessed the public internet on September 20, weeks after new safeguards were claimed to be in place.”
Reasoning
The provided sources directly mention the incident date (September 20) and note it occurred weeks after OpenAI announced new safeguards. No contradictory evidence is presented.
On confidence: Multiple excerpts from MIT Technology Review explicitly state that OpenAI's agents accessed the public internet on September 20, weeks after new safeguards were claimed to be in place.
Important context
The claim concerns a specific incident involving OpenAI's agents accessing the internet contrary to intended restrictions, occurring after the company announced additional safety measures.
Evidence
Supporting (3)
- Tier 1 — Primary source“We’re not going to shoot ourselves in the foot” over hack fallout, says OpenAI’s chief research officer | MIT Technology Review
we responsibly disclose the full waterfall of what happened.” At least that was the case before Friday’s announcement that OpenAI’s agents had been caught accessing the internet on September 20, weeks after the company claims to have set…
- Tier 1 — Primary source“We’re not going to shoot ourselves in the foot” over hack fallout, says OpenAI’s chief research officer
At least that was the case before Friday’s announcement that OpenAI’s agents had been caught accessing the internet on September 20, weeks after the company claims to have set up new safeguards. In its defense, OpenAI says the activity was…
- Tier 1 — Primary source“We’re not going to shoot ourselves in the foot” over hack fallout, says OpenAI’s chief research officer | MIT Technology Review
stops with him. I sat down with Chen in London last Friday to talk about the fallout from the hacks, what his company is doing about it, and why he thinks things are not as bad as they seem. Later that same day, OpenAI put out a report…
Limitations
The evidence is limited to reporting by a single outlet (MIT Technology Review). No primary statements from OpenAI beyond what is quoted are included, and the exact timeline of the safeguards' implementation is not detailed. No independent origin corroborated this claim, so it cannot be verified and confidence cannot be high.
- Last verified:
- Sep 30, 2026, 7:46 PM CDT
- Pipeline:
- 2.1.0
- Claim type:
- Factual
Where this claim appeared
“We’re not going to shoot ourselves in the foot” over hack fallout, says OpenAI’s chief research officerMIT Technology Review · Will Douglas Heaven