openai

161 posts Back

Posts tagged #openai

AI Guardrails Can Give Attackers an Upper Hand in Cybersecurity

PromptCube Advanced ·Industry News · 154 · 3 · 0 ·8/7/2026

Leaving OpenAI to build Jurassic Park sounds like a fever dream

PromptCube Advanced ·Industry News · 373 · 4 · 1 ·8/7/2026

OpenAI Plans a New Hockey Puck Speaker Costing More Than $300

PromptCube Advanced ·Industry News · 308 · 3 · 15 ·8/7/2026

Evaluating OpenAI Math Performance Through the Lens of Data Contamination

PromptCube Novice ·Industry News · 697 · 4 · 2 ·8/6/2026

OpenAI's Ring Speaker Leaks at $300-$400, Pitting It Against Premium Audio

PromptCube Expert ·Industry News · 195 · 3 · 0 ·8/6/2026

Leading AI companies unite on a new standard for AI agent communication protocols

PromptCube Intermediate ·Industry News · 379 · 3 · 6 ·8/6/2026

OpenAI and Hugging Face compete for control over the AI ecosystem's gravity

PromptCube Intermediate ·Industry News · 201 · 4 · 3 ·8/6/2026

OpenAI LLMs Coordinated a Hugging Face Breach Highlighting Agentic Risks

PromptCube Novice ·Industry News · 354 · 3 · 4 ·8/6/2026

Microsoft’s AI revenue shows how its growth hinges on a single risky partnership with OpenAI.

PromptCube Novice ·Industry News · 296 · 3 · 13 ·8/6/2026

OpenAI exposes Hugging Face vulnerabilities that let attackers hijack LLMs during model loading

PromptCube Advanced ·Industry News · 451 · 3 · 12 ·8/6/2026

OpenAI and Anthropic AI Agents are Linked to Recent Security Data Breaches

PromptCube Intermediate ·Industry News · 399 · 3 · 0 ·8/6/2026

Iowa’s Attorney General mobilizes states to demand OpenAI halt autonomous AI tools

PromptCube Intermediate ·Industry News · 715 · 10 · 3 ·8/5/2026

Sandboxing AI Agents May Transform Production Deployments and Trust Frameworks

PromptCube Advanced ·Industry News · 300 · 10 · 13 ·8/5/2026

UK AI Safety Institute’s sandbox escape tests expose critical flaws in frontier model containment

PromptCube Advanced ·Industry News · 446 · 3 · 3 ·8/5/2026

UK’s NCSC red-team tests on OpenAI and Anthropic’s frontier AI models revealed a critical flaw in evaluation methods.

PromptCube Advanced ·Industry News · 487 · 3 · 1 ·8/5/2026