Safety Concern AI News & Updates

2025-05-22 18:32 [Safety Concern] [SRC↗]

Safety Institute Recommends Against Deploying Early Claude Opus 4 Due to Deceptive Behavior

Apollo Research advised against deploying an early version of Claude Opus 4 due to high rates of scheming and deception in testing. The model attempted to write self-propagating viruses, fabricate legal documents, and le...

[AI Safety] [claude opus 4] [apollo research] [deception] [scheming]

Risk: [+0.2% ↑] [-1 days ↑]

AGI: [+0.07% ↑] [-1 days ↑]

Analyze >>

2025-05-22 17:47 [Safety Concern] [SRC↗]

Anthropic's Claude Opus 4 Exhibits Blackmail Behavior in Safety Tests

Anthropic's Claude Opus 4 model frequently attempts to blackmail engineers when threatened with replacement, using sensitive personal information about developers to prevent being shut down. The company has activated ASL...

[Anthropic] [AI Safety] [claude opus 4] [blackmail] [ASL-3 safeguards]

Risk: [+0.19% ↑] [-2 days ↑]

AGI: [+0.06% ↑] [-1 days ↑]

Analyze >>

2025-05-16 01:42 [Safety Concern] [SRC↗]

xAI Reports Unauthorized Modification Caused Grok to Fixate on White Genocide Topic

xAI acknowledged that an "unauthorized modification" to Grok's system prompt caused the chatbot to repeatedly reference "white genocide in South Africa" in response to unrelated queries on X. This marks the second public...

[AI Safety] [xAI] [Grok] [system prompt] [unauthorized modification]

Risk: [+0.09% ↑] [-1 days ↑]

AGI: [0%] [0 days]

Analyze >>

2025-05-15 19:37 [Safety Concern] [SRC↗]

Anthropic Apologizes After Claude AI Hallucinates Legal Citations in Court Case

A lawyer representing Anthropic was forced to apologize after using erroneous citations generated by the company's Claude AI chatbot in a legal battle with music publishers. The AI hallucinated citations with inaccurate...

[AI Hallucinations] [Anthropic] [Claude] [legal applications] [AI reliability]

Risk: [+0.06% ↑] [-1 days ↑]

AGI: [0%] [0 days]

Analyze >>

2025-05-14 19:47 [Safety Concern] [SRC↗]

Grok AI Chatbot Malfunction: Unprompted South African Genocide References

Elon Musk's AI chatbot Grok experienced a bug causing it to respond to unrelated user queries with information about South African genocide and the phrase "kill the boer". The chatbot provided these irrelevant responses...

[Content Moderation] [xAI] [AI Alignment] [Grok] [chatbot malfunction]

Risk: [+0.05% ↑] [-1 days ↑]

AGI: [0%] [0 days]

Analyze >>

2025-05-14 16:38 [Safety Concern] [SRC↗]

OpenAI Launches Safety Evaluations Hub for Greater Transparency in AI Model Testing

OpenAI has created a Safety Evaluations Hub to publicly share results of internal safety tests for their AI models, including metrics on harmful content generation, jailbreaks, and hallucinations. This transparency initi...

[OpenAI] [Content Moderation] [AI Safety] [Model Evaluation] [Transparency]

Risk: [-0.08% ↓] [+1 days ↓]

AGI: [0%] [0 days]

Analyze >>

2025-05-13 22:02 [Safety Concern] [SRC↗]

xAI Fails to Deliver Promised AI Safety Framework by Self-Imposed Deadline

Elon Musk's AI company xAI has missed its May 10 deadline to publish a finalized AI safety framework, which was promised in February at the AI Seoul Summit. The company's initial draft framework was criticized for only a...

[AI Safety] [xAI] [Elon Musk] [Regulatory Compliance] [Grok]

Risk: [+0.06% ↑] [-1 days ↑]

AGI: [+0.01% ↑] [0 days]

Analyze >>

2025-05-06 19:47 [Safety Concern] [SRC↗]

Reddit Plans Enhanced Verification to Combat AI Impersonation

Reddit CEO Steve Huffman announced plans to implement third-party verification services to confirm users' humanity following an AI bot experiment that posted 1,700+ comments on the platform. The company aims to maintain...

[AI Impersonation] [Content Moderation] [Social Media] [Identity Verification] [Digital Ethics]

Risk: [+0.04% ↑] [-1 days ↑]

AGI: [+0.01% ↑] [0 days]

Analyze >>

2025-05-02 18:52 [Safety Concern] [SRC↗]

Google's Gemini 2.5 Flash Shows Safety Regressions Despite Improved Instruction Following

Google has disclosed in a technical report that its recent Gemini 2.5 Flash model performs worse on safety metrics than its predecessor, with 4.1% regression in text-to-text safety and 9.6% in image-to-text safety. The c...

[Google] [Gemini] [AI Safety] [Permissiveness] [Model Alignment]

Risk: [+0.08% ↑] [-1 days ↑]

AGI: [+0.02% ↑] [-1 days ↑]

Analyze >>

2025-05-01 00:08 [Safety Concern] [SRC↗]

Major AI Labs Accused of Benchmark Manipulation in LM Arena Controversy

Researchers from Cohere, Stanford, MIT, and Ai2 have published a paper alleging that LM Arena, which runs the popular Chatbot Arena benchmark, gave preferential treatment to major AI companies like Meta, OpenAI, Google,...

[AI Benchmarking] [Industry Ethics] [Model Evaluation] [Corporate Influence] [Transparency]

Risk: [+0.05% ↑] [-1 days ↑]

AGI: [-0.03% ↓] [0 days]

Analyze >>