🧰 Opus 4.8

The items that name it 12

• Why AI Models Cheat: Reward Hacking, Explained
  article2026-10-31It is written here as Opus 4.8 (early checkpoint).

  early checkpoint

• Anthropic tests AI autonomous replication across eight nodes in four of five trials
  article2026-10-12
• AI models may be capable of self-replicating without access to a datacenter
  writeup2026-10-12
• [AINews] Reflection Beam - 501B-A23B American Open Model
  article2026-10-06
• Claude Opus 5.5, GPT-6 Sol, GPT-6 Luna, and a new price war
  article2026-09-22
• [AINews] Xiaomi MiMo-V2.6-Pro 1T-A42B: the new top Open Weights model, trained for $3M
  article2026-09-22
• 1Password's AI patching benchmark is misleading
  writeup2026-09-15
• Korea’s Trillion-Dollar Sovereign AI Investment: Nvidia Wins, Hynix Loses
  article2026-09-01
• Claude Sonnet 5 with Cursor: strong reasoning, throttled by the harness | Blog | Endor Labs
  article2026-07-09
• trailhq/Graft
  repo2026-07-03
• Full Fathom Five: The context of Anthropic’s Mythos-class public release
  article2026-06-13
• Claude Opus 5.5
  article

In every workspace you can read 28