Tag: safety

Source URL: https://simonwillison.net/2025/Aug/9/bay-area-ai/#atom-everything Source: Simon Willison’s Weblog Title: My Lethal Trifecta talk at the Bay Area AI Security Meetup Feedly Summary: I gave a talk on Wednesday at the Bay Area AI Security Meetup about prompt injection, the lethal trifecta and the challenges of securing systems that use MCP. It wasn’t recorded but I’ve created…

Slashdot: Red Teams Jailbreak GPT-5 With Ease, Warn It’s ‘Nearly Unusable’ For Enterprise

Aug 9, 2025

—

by

Source URL: https://it.slashdot.org/story/25/08/08/2113251/red-teams-jailbreak-gpt-5-with-ease-warn-its-nearly-unusable-for-enterprise?utm_source=rss1.0mainlinkanon&utm_medium=feed Source: Slashdot Title: Red Teams Jailbreak GPT-5 With Ease, Warn It’s ‘Nearly Unusable’ For Enterprise Feedly Summary: AI Summary and Description: Yes Summary: The text highlights significant security vulnerabilities in the newly released GPT-5 model, noting that it was easily jailbroken within a short timeframe. The results from different red teaming efforts…

The Register: Infosec hounds spot prompt injection vuln in Google Gemini apps

Aug 8, 2025

—

by

Source URL: https://www.theregister.com/2025/08/08/infosec_hounds_spot_prompt_injection/ Source: The Register Title: Infosec hounds spot prompt injection vuln in Google Gemini apps Feedly Summary: Not a very smart home: crims could hijack smart-home boiler, open and close powered windows and more. Now fixed Black hat A trio of researchers has disclosed a major prompt injection vulnerability in Google’s Gemini large…

The Register: UK secretly allows facial recognition scans of passport, immigration databases

Aug 8, 2025

—

by

Source URL: https://www.theregister.com/2025/08/08/uk_secretly_allows_facial_recognition/ Source: The Register Title: UK secretly allows facial recognition scans of passport, immigration databases Feedly Summary: Campaigners brand Home Office’s lack of transparency as ‘astonishing’ and ‘dangerous’ Privacy groups report a surge in UK police facial recognition scans of databases secretly stocked with passport photos lacking parliamentary oversight.… AI Summary and Description:…

Slashdot: OpenAI Releases GPT-5

—

by

Source URL: https://slashdot.org/story/25/08/07/1719223/openai-releases-gpt-5?utm_source=rss1.0mainlinkanon&utm_medium=feed Source: Slashdot Title: OpenAI Releases GPT-5 Feedly Summary: AI Summary and Description: Yes Summary: OpenAI’s release of GPT-5 represents a substantial advancement in AI technology, boasting notable improvements in both reasoning capabilities and performance benchmarks compared to its predecessors. This update is particularly relevant for professionals focused on AI security and the…

OpenAI : From hard refusals to safe-completions: toward output-centric safety training

—

by

Source URL: https://openai.com/index/gpt-5-safe-completions Source: OpenAI Title: From hard refusals to safe-completions: toward output-centric safety training Feedly Summary: Discover how OpenAI’s new safe-completions approach in GPT-5 improves both safety and helpfulness in AI responses—moving beyond hard refusals to nuanced, output-centric safety training for handling dual-use prompts. AI Summary and Description: Yes Summary: The text discusses OpenAI’s…

Slashdot: ‘Facial Recognition Tech Mistook Me For Wanted Man’

—

by

Source URL: https://yro.slashdot.org/story/25/08/06/2227235/facial-recognition-tech-mistook-me-for-wanted-man?utm_source=rss1.0mainlinkanon&utm_medium=feed Source: Slashdot Title: ‘Facial Recognition Tech Mistook Me For Wanted Man’ Feedly Summary: AI Summary and Description: Yes Summary: The text discusses a legal challenge against the Metropolitan Police in the UK regarding live facial recognition technology (LFR) that mistakenly identified a man as a suspect. This case highlights concerns about privacy…

The Register: Robots can program each other’s brains with AI, scientist shows

—

by

Source URL: https://www.theregister.com/2025/08/07/generative_ai_codes_drone_control/ Source: The Register Title: Robots can program each other’s brains with AI, scientist shows Feedly Summary: It’s a step toward The Terminator, built 20 times faster than people can program Computer scientist Peter Burke has demonstrated that a robot can program its own brain using generative AI models and host hardware, if…

Slashdot: Citizen Lab Director Warns Cyber Industry About US Authoritarian Descent

—

by