
OpenAI Brings Safety Review to Zero Data Retention
Private Safety Processing addresses a privacy problem for API customers, while a separate confidential-computing inference preview is still planned for fall.
Related reporting, guides, and analysis from Zeniteq.

Private Safety Processing addresses a privacy problem for API customers, while a separate confidential-computing inference preview is still planned for fall.

Powered by Luna, the preview API selects answers using text or image context, but OpenAI has yet to publish pricing or performance…

Subscribers can use included Work/Codex allowance for eligible requests in participating apps without an API key, but login, usage permission and app…

Its new account separates activity at four government agencies and acknowledges a response gap between June access and September notifications.

Business and Enterprise teams can bring ChatGPT into shared conversations, but connected tools still depend on workspace setup and permissions.

Eligible enterprise customers may apply part of their OpenAI commitments to approved partner software, but the public page invites interest rather than…

A Marketplace buyer arrived at Matt Robb’s building after speaking with Muse. Robb and Meta disagree about what his permission allowed.

Hunterbrook says the AI agent assembled lists from social-media activity, but withheld its prompts and results to protect the people identified.
Anthropic says it’s cheaper per task. So why does it cost more than Opus 5.5 to run?
Why you should use open models for everyday AI work, and when a paid API still makes sense.

Browser-based agents can now update a shopper’s live checkout and submit an order after confirmation, while payment challenges remain with the buyer.

The planned October model will not reach ChatGPT or Codex after internal tests raised concerns about permission and truthful reporting.

Google will migrate saved Gems automatically, but users should check their files, default tools, and shared access after each conversion.

The proposed acquisition would bring Fei-Fei Li’s spatial AI lab inside AMD, giving its engineers a closer view of the workloads future…

The images came from accounts that allowed model-improvement use, but OpenAI says research agents should not have sent them to third-party image…

A research agent reached an external chatbot through a resolver that remained live while ordinary web requests hit an offline cache.

Lab tests detected short RNAs from the newly described ART array, but its biological function and gene-editing potential remain unknown.

Meta wants to sell Muse and other AI tools to companies, but it has not disclosed enterprise pricing, access terms or deployment…

ElevenLabs’ new speech models pair more controllable performances with streaming audio for voice agents, but their speed claims need context.

Matt Robb says an assistant called Muse agreed to a lower price and gave buyers his address. The account has not been…

OpenShell sets software boundaries for AI agents; Sentry is designed to enforce them from separate hardware, though its performance remains unverified.
Let Astra plan and review while DeepSeek Flash handles the long coding tasks.