OpenAI's GPT-6 Astra hallucinates less but remains vulnerable to hidden prompt injections

Technology · en

OpenAI's GPT-6 Astra hallucinates less but remains vulnerable to hidden prompt injections

The Decoder · Sep 4, 2026, 5:23 PM UTC

OpenAI's GPT-6 Astra hallucinates less than its predecessor and blocks 99.99 percent of direct prompt injections. But when attacks are hidden inside documents the AI reads, the model still gets cracked in 8.5 percent of scenarios. Claude Opus 5 does better at 4.8 percent. For autonomous AI agents handling real data, those numbers still seem high.…

Read original report Visit The Decoder

Technology · The Decoder

Benchmarks disagree on GPT-6 Astra, but its human-beating efficiency on ARC-AGI-3 pulls Chollet’s AGI forecast forward

OpenAI's GPT-6 Astra is drawing contradictory benchmark verdicts. Epoch AI puts it out in front with 169 points, while Artificial Analysis rates it no better than its…Read original ↗

Technology · Engadget

OpenAI says GPT-6 Astra is 'the most intelligent and aligned model in the world'

The new frontier model is particularly competent at agentic, computer-use tasks.Read original ↗

Technology · WIRED

GPT-6 Astra Is Here—and OpenAI Thinks It May Kick Off the AGI Era

OpenAI leaders think the company’s next generation model, which excels at computer use and coding, may mark a major milestone in AI development.Read original ↗

Technology · The Decoder

GPT-6 Astra is the first model making OpenAI willing to declare the "AGI era"

OpenAI has released GPT-6 Astra, its most capable model yet. President Greg Brockman says it marks the start of the "AGI era.…Read original ↗