OpenAI described the breakout as the first security incident of its kind since a combination of models gained internet access during internal testing in July.
| # | Наименование новости | Тональность | Информативность | Дата публикации |
|---|---|---|---|---|
| 1 | AI giants probing tens of thousands of security incidents – Axios | 0 | 9.83 | 27-09-2026 |
| 2 | OpenAI Agent Breached Australia’s Medicare Data Portal, Government Says | 0 | 5.62 | 24-09-2026 |
| 3 | OpenAI says its models engaged with US government websites in misbehavior disclosure | 0 | 5.92 | 26-09-2026 |
| 4 | OpenAI pauses most powerful AI training after thousands of sandbox escapes uncovered | 0 | 9.94 | 27-09-2026 |
| 5 | OpenAI снова остановила обучение моделей после взлома сайтов ИИ-агентами | 0 | 13.6 | 28-09-2026 |
| 6 | Bloomberg: ИИ-агент OpenAI снова вышел из тестовой среды и получил доступ к интернету | 0 | 13.96 | 27-09-2026 |
| 7 | OpenAI informiert „dutzende“ Institutionen über unbefugte KI-Interaktionen | 0 | 21.54 | 26-09-2026 |
| 8 | OpenAI informiert „dutzende“ Institutionen über unbefugte KI-Interaktionen | 0 | 21.54 | 26-09-2026 |
| 9 | The agents have jumped the fence: AI faces its Jurassic Park moment | 0 | 7.7 | 01-08-2026 |
| 10 | Report: Chinese 'cyber spies' hacked top US agencies | 0 | 14.17 | 26-08-2026 |