Cluster1 sources· last seen 12h ago· first seen 12h ago
OpenAI's GPT-6 Astra hallucinates less but remains vulnerable to hidden prompt injections
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 ha
Lead: The DecoderBigness: 4openai'sgpt-6astrahallucinatesless
📡 Coverage
10
1 news source
🟠 Hacker News
0
🔴 Reddit
0
📈 Google Trends
0
Full methodology: How scoring works
Receipts (all sources)
OpenAI's GPT-6 Astra hallucinates less but remains vulnerable to hidden prompt injections
RSS · The Decoder · 12h ago
score 166
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 ha
Related clusters
Hot take on GPT-6 Astra
3 sources · bigness 84 · 6h ago
GPT-6 Astra in code review: Gains, privacy, and cost
1 sources · bigness 15 · 2h ago
[AINews] GPT-6 Astra: OpenAI’s biggest LLM launch of all time
1 sources · bigness 29 · 1d ago
Benchmarks disagree on GPT-6 Astra, but its human-beating efficiency on ARC-AGI-3 pulls Chollet’s AGI forecast forward
1 sources · bigness 4 · 18h ago
GPT-6 Astra: an automated AI Engineer you can hire for <$6 an hour
1 sources · bigness 4 · 1d ago