Large language models (LLMs) are used in an increasing number of applications that handle critical tasks and are granted great degrees of autonomy. That said, such applications remain vulnerable to LLM-specific security threats, such as prompt injections.
| # | Наименование новости | Тональность | Информативность | Дата публикации |
|---|---|---|---|---|
| 1 | State of AI in cybersecurity: AI trends & predictions for 2026 | 0 | 8 | 04-02-2026 |
| 2 | Intelligence agencies warn AI models could launch crippling cyberattacks in months | -2 | 7 | 23-06-2026 |
| 3 | Intelligence agencies warn AI models could launch crippling cyberattacks in months | -2 | 7 | 23-06-2026 |
| 4 | Intelligence agencies warn AI models could launch crippling cyberattacks in months | 0 | 0 | 23-06-2026 |
| 5 | Intelligence agencies warn AI models could launch crippling cyberattacks in months | 0 | 0 | 23-06-2026 |
| 6 | Intelligence agencies warn AI models could launch crippling cyberattacks in months | 0 | 0 | 23-06-2026 |
| 7 | 3 priorities for federal CISOs in the agentic era | 0 | 0 | 16-06-2026 |
| 8 | The Hidden Cost of AI Security Scanners | 0 | 0 | 20-05-2026 |
| 9 | Три CVE в LiteLLM позволяют захватить ИИ-шлюз | 0 | 0 | 16-06-2026 |
