Submissions from futuresearch.ai

		AI takes people at their word (futuresearch.ai)
		5 points by mckennameyer 4 days ago \| past \| 1 comment
		Claude can miss the motives of politicians (futuresearch.ai)
		10 points by ddp26 9 days ago \| past \| discuss
		Measuring one way AIs lack self-awareness (futuresearch.ai)
		1 point by ddp26 15 days ago \| past
		How long until AI automates all cognitive labor? (futuresearch.ai)
		45 points by mckennameyer 16 days ago \| past \| 82 comments
		Some rare examples of AIs being underconfident (futuresearch.ai)
		6 points by ddp26 18 days ago \| past
		History doesn't repeat itself as often as LLMs think (futuresearch.ai)
		1 point by ddp26 22 days ago \| past
		Agents Sometimes Catastrophize (futuresearch.ai)
		9 points by ddp26 22 days ago \| past \| 2 comments
		Apple's Plan to Power Siri with ChatGPT Was a Predictable Failure (futuresearch.ai)
		3 points by mckennameyer 29 days ago \| past
		Run Agents Twice (futuresearch.ai)
		6 points by ddp26 31 days ago \| past
		Claude is that gullible friend who takes everyone at their word (futuresearch.ai)
		1 point by mckennameyer 31 days ago \| past
		Estimating the market value of top AI researchers (futuresearch.ai)
		11 points by Bullhorn9268 58 days ago \| past
		I think Anthropic is worth $100B more than last week (futuresearch.ai)
		9 points by ddp26 65 days ago \| past
		A forecast of the fair market value of SpaceX's businesses (futuresearch.ai)
		100 points by ddp26 71 days ago \| past \| 205 comments
		My minute-by-minute response to the LiteLLM malware attack (futuresearch.ai)
		441 points by Fibonar 79 days ago \| past \| 157 comments
		Supply Chain Attack in litellm 1.82.8 on PyPI (futuresearch.ai)
		8 points by vnorilo 80 days ago \| past \| 1 comment
		LiteLLM PyPI has been compromised an hour ago, do not update (futuresearch.ai)
		27 points by Bullhorn9268 81 days ago \| past \| 5 comments
		`new Date()` Considered Harmful (futuresearch.ai)
		7 points by rgambee 87 days ago \| past
		Claude Code's MCP config can silently orphan Docker containers (futuresearch.ai)
		8 points by rgambee 89 days ago \| past \| 2 comments
		How Predictable Are the Oscars? (futuresearch.ai)
		10 points by nbosse 3 months ago \| past \| 2 comments
		LLMs Are Finally Good Enough to Analyse Their Own Traces (futuresearch.ai)
		7 points by rgambee 3 months ago \| past
		Inline MCP results are the new prompt bloat (futuresearch.ai)
		9 points by rafaelpo 3 months ago \| past \| 2 comments
		Ask LLM Agents to Classify Problems Before Starting (futuresearch.ai)
		7 points by ddp26 3 months ago \| past \| 1 comment
		How AI Finds Fuzzy Duplicates in Large Datasets (futuresearch.ai)
		10 points by nbosse 3 months ago \| past \| 1 comment
		Higher effort reduces deep research accuracy for Gemini Flash 3 and GPT-5 (futuresearch.ai)
		11 points by wawawildwildest 4 months ago \| past \| 1 comment
		How LLM agents solve the table merging problem (futuresearch.ai)
		29 points by ddp26 4 months ago \| past \| 3 comments
		Forecasting the 2026 AI Winner (futuresearch.ai)
		2 points by ddp26 5 months ago \| past \| 1 comment
		[flagged] The Karpathy Interview, 6 Months After AI 2027 (futuresearch.ai)
		38 points by ddp26 7 months ago \| past \| 24 comments
		Unit Economics of LLM APIs (futuresearch.ai)
		3 points by supriyo-biswas on March 27, 2025 \| past
		Apple's plan to power Siri with ChatGPT was a predictable failure (futuresearch.ai)
		15 points by ddp26 on March 10, 2025 \| past \| 1 comment
		OpenAI Deep Research – Six Strange Failures (futuresearch.ai)
		8 points by ddp26 on Feb 19, 2025 \| past
		More

HN For You