OpenAI launches new Astra model amid growing scrutiny over agents' safety
Economic Times Techen
The concerns center on agentic AI, which is designed to perform tasks with little to no human intervention. The promise of agents running around the clock is central to investors' confidence in the appeal of AI as a transformative technology.
This is a short summary published by AI Global Wire. The full article is owned and hosted by Economic Times Tech — open it there to read it in full.
Read the full story at Economic Times Tech- OpenAI
- Verktyg
- Agenter
Related AI news
- Analysis of Prompt Engineering for Drug Toxicity PredictionarXiv cs.AI · September 4, 2026
- Structure and Implementation of New Practical English Textbooks Driven by Artificial IntelligencearXiv cs.AI · September 4, 2026
- Speculative Macro Commit for Faster Tool-Using AgentsarXiv cs.AI · September 4, 2026
- Fresh Memory, Stale Plans: Dependency-Scoped Validation for Distributed LLM-Agent MemoryarXiv cs.AI · September 4, 2026
- A Prompt-Engineering Approach to Develop Scalable, Flexible, and Real-Time Hybrid Micro-Level Personalization in a General Purpose AI Teaching AssistantarXiv cs.AI · September 4, 2026
- DuplexSpeechBench-IFEval: Evaluating Implicit Instruction Following in Full-Duplex Voice AgentsarXiv cs.AI · September 4, 2026