New Grok 3 release tops LLM leaderboards despite Musk-approved “based” opinions
AI expert Andrej Karpathy tested Grok 3 and wrote on X, "As far as a…
Privacy-problematic DeepSeek pulled from app stores in South Korea
In a media briefing held Monday, the South Korean Personal Information Protection Commission indicated that…
Posted inAI Features generative ai
Reddit mods are fighting to keep AI slop off subreddits. They could use help.
Some moderators say generative AI helps people spam unwanted content on a subreddit, including posts…
Posted inAI AI assistants AI paternalism
ChatGPT can now write erotica as OpenAI eases up on AI paternalism
"Following the initial release of the Model Spec (May 2024), many users and developers expressed…
Posted inAI biochemistry Biology
AI used to design a multi-step enzyme that can digest some plastics
And it worked. Repeating the same process with an added PLACER screening step boosted the…
Posted inAI Artificial Intelligence Conde Nast
Condé Nast, other news orgs say AI firm stole articles, spit out “hallucinations”
Condé Nast and several other media companies sued the AI startup Cohere today, alleging that…
Posted inAI
Over half of LLM-written news summaries have “significant issues”—BBC analysis
Here at Ars, we've done plenty of coverage of the errors and inaccuracies that LLMs…
Posted inAI
Sam Altman lays out roadmap for OpenAI’s long-awaited GPT-5 model
In March 2024, for example, Altman said in an interview with Lex Friedman that the…
Posted inAI Artificial Intelligence Biz & IT
New hack uses prompt injection to corrupt Gemini’s long-term memory
Google Gemini: Hacking Memories with Prompt Injection and Delayed Tool Invocation. Based on lessons learned…
Posted inAI chrome chrome browser
Google Chrome may soon use “AI” to replace compromised passwords
Google's Chrome browser might soon get a useful security upgrade: detecting passwords used in data…