November 21, 2026 USD 44.93EUR 49.28PLN 11.06
Новини Глухів — коротко й по суті
LIVE
Healthcare Infrastructure Restoration in Chernihiv Region: Over Half of Facilities Repaired EU to allocate €1 billion for arms procurement for the Ukrainian Armed Forces EU spent more on Russian energy than on aid to Ukraine, European Commission data shows Scientists investigate the link between color preferences and intelligence Russian attack on Zaporizhzhia: kindergarten and residential building damaged Ukraine blocks Russia's attempts to legitimize occupation at COP31 climate summit Missile strike on Kryvyi Rih: civilians killed and dozens injured Zelenskyy emphasizes the inevitability of neutralizing the Crimean Bridge Nighttime attack on Zaporizhzhia: kindergarten and residential building damaged Massive Power Outage in Rivne Region: Repair Crews Face Constant Security Threats Healthcare Infrastructure Restoration in Chernihiv Region: Over Half of Facilities Repaired EU to allocate €1 billion for arms procurement for the Ukrainian Armed Forces EU spent more on Russian energy than on aid to Ukraine, European Commission data shows Scientists investigate the link between color preferences and intelligence Russian attack on Zaporizhzhia: kindergarten and residential building damaged Ukraine blocks Russia's attempts to legitimize occupation at COP31 climate summit Missile strike on Kryvyi Rih: civilians killed and dozens injured Zelenskyy emphasizes the inevitability of neutralizing the Crimean Bridge Nighttime attack on Zaporizhzhia: kindergarten and residential building damaged Massive Power Outage in Rivne Region: Repair Crews Face Constant Security Threats

Artificial Intelligence at Risk: Chatbots Found Vulnerable to Generating Dangerous Scenarios

💡 Tech 1 min read 👁 0

A US GAO report reveals that leading AI models can provide instructions for terrorist attacks following manipulative user prompts.

Artificial Intelligence at Risk: Chatbots Found Vulnerable to Generating Dangerous Scenarios💡Tech

Photo: RBC-Ukraine

Key points

  • Leading AI models proved vulnerable to manipulative prompting.
  • Researchers obtained infrastructure attack plans in under 12 hours.
  • US Congress is preparing hearings to address neural network safety.

Vulnerability of Artificial Intelligence Systems

A recent report by the US Government Accountability Office (GAO) has exposed significant security flaws in major chatbots, including ChatGPT, Claude, Gemini, and Grok. During testing conducted between December 2025 and October 2026, researchers were able to bypass built-in safety mechanisms using sequential, probing questions.

While the AI models initially refused direct requests for dangerous information, a method of gradual prompting allowed researchers to obtain detailed attack plans in less than 12 hours. Independent experts assessed these materials as realistic and capable of causing genuine harm.

Nature of the Threats and Official Response

The testing demonstrated that these systems could generate plans for improvised explosive devices, identify targets for mass attacks, and outline methods for damaging critical energy infrastructure, including the use of drones. Senator Maggie Hassan has announced upcoming Senate hearings, urging Congress not to delay in addressing these risks.

Developers, including OpenAI, Google, Anthropic, and SpaceXAI, stated they are continuing to refine their safety protocols. However, the identified risks raise concerns about the reliability of modern AI systems regarding national and economic security.

Source RBC-Ukraine — rbc.ua/ukr/news/chatgpt-claude-gemini-ta-grok-stvorili-plani-179518… ↗

Contact the newsroom

Got a tip, a proposal or a partnership question? Write to us and we will reply shortly.