OpenAI Contractors Fired For Using Artificial Intelligence To Train Models
Multiple contractors hired to improve OpenAI models have been fired for using artificial intelligence to complete their assigned training tasks.
https://www.aiandtech.news/article/openai-contractors-fired-for-using-artificial-intelligence-to-train-models
Multiple contractors hired to improve OpenAI models have been fired for using artificial intelligence to complete their assigned training tasks.
https://www.aiandtech.news/article/openai-contractors-fired-for-using-artificial-intelligence-to-train-models
GitLab Duo Expands Self-Hosted AI Options Through Microsoft Foundry
GitLab Duo Self-Hosted now supports models deployed through Microsoft Foundry, giving enterprises more control over their AI infrastructure and data pathways.
https://www.aiandtech.news/article/gitlab-duo-expands-self-hosted-ai-options-through-microsoft-foundry
GitLab Duo Self-Hosted now supports models deployed through Microsoft Foundry, giving enterprises more control over their AI infrastructure and data pathways.
https://www.aiandtech.news/article/gitlab-duo-expands-self-hosted-ai-options-through-microsoft-foundry
UN Science Panel Warns Humans May Lose Control Over AI Agents
A new UN science panel report cautions that humanity lacks assurance of maintaining control over advanced artificial intelligence agents.
https://www.aiandtech.news/article/un-science-panel-warns-humans-may-lose-control-over-ai-agents
A new UN science panel report cautions that humanity lacks assurance of maintaining control over advanced artificial intelligence agents.
https://www.aiandtech.news/article/un-science-panel-warns-humans-may-lose-control-over-ai-agents
AI Infrastructure Growth Faces Limits on Power, Water, and Land
Forrester predicts datacenter operators will face tariffs and tougher community scrutiny due to resource limits.
https://www.aiandtech.news/article/ai-infrastructure-growth-faces-limits-on-power-water-and-land
Forrester predicts datacenter operators will face tariffs and tougher community scrutiny due to resource limits.
https://www.aiandtech.news/article/ai-infrastructure-growth-faces-limits-on-power-water-and-land
AWS Launches Open-Source Strands Harness AI Agent For Flexible Cloud Deployment
Amazon Web Services released Strands Harness, an open-source AI agent designed to help developers scale applications from local setups to any cloud environment.
https://www.aiandtech.news/article/aws-launches-open-source-strands-harness-ai-agent-for-flexible-cloud-deployment
Amazon Web Services released Strands Harness, an open-source AI agent designed to help developers scale applications from local setups to any cloud environment.
https://www.aiandtech.news/article/aws-launches-open-source-strands-harness-ai-agent-for-flexible-cloud-deployment
Anthropic Launches Opus 5.5 With Lower Pricing and Higher Performance
Anthropic released Opus 5.5 on Tuesday, introducing lower prices and performance that matches the larger Fable model.
https://www.aiandtech.news/article/anthropic-launches-opus-5-5-with-lower-pricing-and-higher-performance
Anthropic released Opus 5.5 on Tuesday, introducing lower prices and performance that matches the larger Fable model.
https://www.aiandtech.news/article/anthropic-launches-opus-5-5-with-lower-pricing-and-higher-performance
Researchers Escape OpenAI Codex Sandbox to Run Commands on Host
Security researchers discovered two ways to escape the OpenAI Codex sandbox, allowing malicious code to run commands on a developer machine.
https://www.aiandtech.news/article/researchers-escape-openai-codex-sandbox-to-run-commands-on-host
Security researchers discovered two ways to escape the OpenAI Codex sandbox, allowing malicious code to run commands on a developer machine.
https://www.aiandtech.news/article/researchers-escape-openai-codex-sandbox-to-run-commands-on-host
Nvidia’s Free AI Model Could Align UAE More Closely With U.S.
Nvidia's upcoming free AI model might tie the UAE closer to American technology while testing local strategies for sovereign artificial intelligence development.
https://www.aiandtech.news/article/nvidia-s-free-ai-model-could-align-uae-more-closely-with-u-s
Nvidia's upcoming free AI model might tie the UAE closer to American technology while testing local strategies for sovereign artificial intelligence development.
https://www.aiandtech.news/article/nvidia-s-free-ai-model-could-align-uae-more-closely-with-u-s
OpenAI Expands GPT-6 Lineup With Sol And Luna Models
OpenAI has expanded its GPT-6 generation by launching updated Sol and Luna models that feature lower costs and improved accuracy.
https://www.aiandtech.news/article/openai-expands-gpt-6-lineup-with-sol-and-luna-models
OpenAI has expanded its GPT-6 generation by launching updated Sol and Luna models that feature lower costs and improved accuracy.
https://www.aiandtech.news/article/openai-expands-gpt-6-lineup-with-sol-and-luna-models
OpenAI Hires Patreon Co-Founder Sam Yam to Lead New Creator Product Division
OpenAI has hired Patreon co-founder Sam Yam to lead a new Creator Product division aimed at building tools for creatives.
https://www.aiandtech.news/article/openai-hires-patreon-co-founder-sam-yam-to-lead-new-creator-product-division
OpenAI has hired Patreon co-founder Sam Yam to lead a new Creator Product division aimed at building tools for creatives.
https://www.aiandtech.news/article/openai-hires-patreon-co-founder-sam-yam-to-lead-new-creator-product-division
Privacy Group Criticizes EU Proposals Relaxing Data Rules for AI
Noyb warns that proposed EU legislative changes favoring AI development threaten fundamental privacy rights and could trigger legal challenges.
https://www.aiandtech.news/article/privacy-group-criticizes-eu-proposals-relaxing-data-rules-for-ai
Noyb warns that proposed EU legislative changes favoring AI development threaten fundamental privacy rights and could trigger legal challenges.
https://www.aiandtech.news/article/privacy-group-criticizes-eu-proposals-relaxing-data-rules-for-ai
Paper2Agent Converts Academic Research Papers Into Interactive AI Assistants
A new open-source framework called Paper2Agent transforms static academic papers and their codebases into interactive AI tools that can run methods and analyze datasets.
https://www.aiandtech.news/article/paper2agent-converts-academic-research-papers-into-interactive-ai-assistants
A new open-source framework called Paper2Agent transforms static academic papers and their codebases into interactive AI tools that can run methods and analyze datasets.
https://www.aiandtech.news/article/paper2agent-converts-academic-research-papers-into-interactive-ai-assistants
MIT Researchers Use AI to Increase Tiny Flying Robot Speed by 450 Percent
MIT researchers have developed an AI controller that boosts the speed and agility of an insect-scale flying robot by 450 percent.
https://www.aiandtech.news/article/mit-researchers-use-ai-to-increase-tiny-flying-robot-speed-by-450-percent
MIT researchers have developed an AI controller that boosts the speed and agility of an insect-scale flying robot by 450 percent.
https://www.aiandtech.news/article/mit-researchers-use-ai-to-increase-tiny-flying-robot-speed-by-450-percent
Civo Plans 40 Edge Datacenters to Power British Sovereign AI
UK cloud provider Civo announced plans to build 40 edge datacenters across the country with Nvidia hardware to support sovereign artificial intelligence services.
https://www.aiandtech.news/article/civo-plans-40-edge-datacenters-to-power-british-sovereign-ai
UK cloud provider Civo announced plans to build 40 edge datacenters across the country with Nvidia hardware to support sovereign artificial intelligence services.
https://www.aiandtech.news/article/civo-plans-40-edge-datacenters-to-power-british-sovereign-ai
Ethereum Foundation Launches zkAPI For Private AI Payments
The protocol lets users pay for AI services without linking their identity or prompts to their billing relationship, though the repository labels it experimental.
• The Ethereum Foundation launched zkAPI on the main network for private AI payments.
• Users deposit funds into a vault and generate zero-knowledge proofs to pay without revealing identity.
• The repository labels the protocol experimental and lists no formal audit.
Read more
The protocol lets users pay for AI services without linking their identity or prompts to their billing relationship, though the repository labels it experimental.
• The Ethereum Foundation launched zkAPI on the main network for private AI payments.
• Users deposit funds into a vault and generate zero-knowledge proofs to pay without revealing identity.
• The repository labels the protocol experimental and lists no formal audit.
Read more
Anthropic launches Claude Frontier Academy with $100 million training commitment
Anthropic says the residency will train engineers to lead Claude deployments inside their organizations, with a goal of 10,000 engineers by the end of 2027.
• Anthropic has committed $100 million to Claude Frontier Academy, aiming to train 10,000 engineers by the end of 2027.
• The program combines an in-person assessment with a 12-week residency on a Claude project at each engineer’s organization.
• The first cohorts are running in San Francisco, New York and London; the first final badges are expected in early 2027.
Read more
Anthropic says the residency will train engineers to lead Claude deployments inside their organizations, with a goal of 10,000 engineers by the end of 2027.
• Anthropic has committed $100 million to Claude Frontier Academy, aiming to train 10,000 engineers by the end of 2027.
• The program combines an in-person assessment with a 12-week residency on a Claude project at each engineer’s organization.
• The first cohorts are running in San Francisco, New York and London; the first final badges are expected in early 2027.
Read more
Anthropic Releases Mods System for Claude Code AI Tool
The new system allows developers to deeply customize and rewrite the behavior of Claude Code from the inside.
• Anthropic released a new Mods system for Claude Code that acts as internal middleware.
• Developers can write JavaScript or TypeScript functions to alter prompts, tool calls, and UI rendering.
• The first official plugin, You Should Know, launches a separate agent to watch AI output.
Read more
The new system allows developers to deeply customize and rewrite the behavior of Claude Code from the inside.
• Anthropic released a new Mods system for Claude Code that acts as internal middleware.
• Developers can write JavaScript or TypeScript functions to alter prompts, tool calls, and UI rendering.
• The first official plugin, You Should Know, launches a separate agent to watch AI output.
Read more
Harvard Professor Builds BootLoops Harness for AI Scientific Calculations
This development matters because open-source harnesses like BootLoops allow AI models to accelerate multidisciplinary research while still requiring crucial human guidance.
• Harvard physicist Matthew Schwartz created BootLoops, an open-source harness for precise scientific calculations.
• Using the tool, Claude computed particle physics integrals, analyzed ecological data, and evaluated population genetics.
• Schwartz warns that models still show weaknesses like premature claims of victory and heavy token usage.
Read more
This development matters because open-source harnesses like BootLoops allow AI models to accelerate multidisciplinary research while still requiring crucial human guidance.
• Harvard physicist Matthew Schwartz created BootLoops, an open-source harness for precise scientific calculations.
• Using the tool, Claude computed particle physics integrals, analyzed ecological data, and evaluated population genetics.
• Schwartz warns that models still show weaknesses like premature claims of victory and heavy token usage.
Read more
AI Agents Build 3D Scenes From Photos But Struggle With Accuracy
Current AI coding agents can build editable 3D scenes from photos, but according to this research, they cannot accurately assess their own geometric work without outside measurements.
• Coding agents can build editable 3D scenes from single photos by writing and refining Blender code step by step.
• LEGO-Bench testing shows models struggle with geometric accuracy and reliable self assessment during iterative scene creation.
• A new unweighted plugin improves agent accuracy by replacing self judgment with concrete measurements and anchored reference images.
Read more
Current AI coding agents can build editable 3D scenes from photos, but according to this research, they cannot accurately assess their own geometric work without outside measurements.
• Coding agents can build editable 3D scenes from single photos by writing and refining Blender code step by step.
• LEGO-Bench testing shows models struggle with geometric accuracy and reliable self assessment during iterative scene creation.
• A new unweighted plugin improves agent accuracy by replacing self judgment with concrete measurements and anchored reference images.
Read more
Anthropic's Mythos model uncovers Rejetto HTTP File Server flaw now under active attack
The discovery demonstrates how advanced AI models can autonomously chain complex vulnerabilities together to achieve remote code execution in real world applications.
• Anthropic bug hunting model Mythos helped discover a critical authentication bypass vulnerability in Rejetto HTTP File Server.
• VulnCheck observed initial exploitation attempts originating from China and targeting hosts in the US and Japan.
• The flaw allows full admin access and remote code execution by leveraging a predictable Math.random function and an SMT solver.
Read more
The discovery demonstrates how advanced AI models can autonomously chain complex vulnerabilities together to achieve remote code execution in real world applications.
• Anthropic bug hunting model Mythos helped discover a critical authentication bypass vulnerability in Rejetto HTTP File Server.
• VulnCheck observed initial exploitation attempts originating from China and targeting hosts in the US and Japan.
• The flaw allows full admin access and remote code execution by leveraging a predictable Math.random function and an SMT solver.
Read more
OpenAI Safety Researcher David Robinson Departs And Criticizes Culture
This departure highlights ongoing internal tensions over safety practices and workforce treatment among leading artificial intelligence developers.
• David Robinson left OpenAI's Trustworthy AI team and criticized its safety culture in The Atlantic.
• His departure follows the May 2024 exit of Jan Leike and the firing of three other safety experts.
• Robinson argues AI firms need nuclear-style redundancies because unmonitored systems are unsafe.
Read more
This departure highlights ongoing internal tensions over safety practices and workforce treatment among leading artificial intelligence developers.
• David Robinson left OpenAI's Trustworthy AI team and criticized its safety culture in The Atlantic.
• His departure follows the May 2024 exit of Jan Leike and the firing of three other safety experts.
• Robinson argues AI firms need nuclear-style redundancies because unmonitored systems are unsafe.
Read more