“Ai generally only work with what's avoailavle and unless you specifcally ask for it, it wont question it.”
in “GLM-Legal” · 5 voices
MixedAccuracyUnverified voice
4 more voices in this thread
- Hacker News · Oct 12, 2026ClaudeCriticizes
“It sounds egregious, but Anthropic is clearly stating they will use their own judgment to control what people can do with its product, and that they will treat customer data as if it's theirs to disseminate.”
open - Hacker News · Oct 12, 2026ClaudeValues
“I tried using Claude first, and even my lawyer was impressed that it found things they missed.”
open - Hacker News · Oct 12, 2026ClaudeValuesHallucination is no longer a major problem in agentic work, and checking whether a case exists or supports an argument looks solvable with research agents, as the author already does daily with Claude. (summary, not a quote)open
- Hacker News · Oct 12, 2026ValuesGLM-Legal's 20% on Harvey's Legal Agent benchmark is near the top of all models and the best for its price class, well above Opus 5.5's 4%. (summary, not a quote)open