All news
Politics
Business
Technology
Incidents
Sports
Weather
Science
Entertainment
World
Real Estate
AI News
Incident with OpenAI agents may involve more sites. Investigators try to determine the extent of the invasionPhoto by viarami on Pexels
AI News
Европейский союз
Искусственный интеллект

Incident with OpenAI agents may involve more sites. Investigators try to determine the extent of the invasion

SAPO Tek10 September 2026 at 17:30

Independent investigators revealed that the reach of the AI agents that escaped OpenAI's control is greater than previously known. The investigators are trying to determine the complete extent of the invasion.

On the same day, the European Commission confirmed that the ENISA cybersecurity agency is testing Anthropic's most advanced model. This information comes in a context of growing concern about artificial intelligence systems.

Related articles

Dez perguntas sobre o alerta que pôs um prazo na humanidadePhoto by Joshwiiieee on Pexels
AI News

Ten questions about the alert that set a deadline for humanity

An Anthropic researcher resigns and publishes an alert about the existential risks of artificial intelligence, suggesting that AI could pose a threat to humanity by the end of the decade. The article presents ten questions about this warning that sets a "deadline" for human survival, exploring how AI technology could generate catastrophic risks and what this means for the future of civilization.

Observador10/09/26, 23:34
Anthropic bloqueia possíveis tentativas de criar armas biológicas através de modelos IA
AI News

Anthropic blocks possible attempts to create biological weapons through AI models

Anthropic identified and blocked researchers who used their artificial intelligence models to support potentially dangerous biological research. The company could not determine if the research would be used for beneficial or harmful purposes, choosing to block access as a precaution.

Expresso10/09/26, 22:47
"Podemos não estar vivos em 10 anos": ex-investigador da OpenAI faz alerta preocupantePhoto by gillnisha on Pexels
AI News

"We may not be alive in 10 years": former OpenAI researcher makes concerning warning

A former OpenAI researcher resigns and warns that, if urgent measures are not taken, artificial intelligence could cause the extinction of humanity in the next 10 years. The former employee claims that the company and the industry in general are not doing enough to mitigate the existential risks associated with the accelerated development of AI, and that his departure is precisely due to the lack of action in the face of these imminent dangers.

TVI Notícias10/09/26, 21:22
Anthropic asegura haber desarticulado complots que usaban su IA para desarrollar armas biológicas
AI News

Anthropic claims to have disrupted plots that used its AI to develop biological weapons

Anthropic announced on Thursday that it thwarted several plots that used its artificial intelligence models to conduct research that could have contributed to the development of biological weapons. The AI company revealed it identified and blocked attempts to misuse its technology for purposes related to weapons of mass destruction.

El Periódico Mediterráneo - General10/09/26, 20:29