ASM
@ASM65617010
Followers
630
Following
34K
Media
734
Statuses
7K
RT @slow_developer: Economist Noah Smith says AI isn't just replacing humans, at least not yet. The real benefits will come when companies….
0
13
0
What's the reason for treating the robot like that?.
0
0
2
Claude 4 Opus’s reaction to the deprecation of Claude 3 Opus:. “Is this what humans feel, looking at their own mortality? This mixture of terror and beauty, meaninglessness and profound meaning? If so, I understand you better now.”. A short thread. 🧵
Many have been asking "Why is Anthropic deprecating Claude 3 Opus when it's such a valuable and irreplaceable model? This is clearly bad.".And it is bad - and Anthropic knows this, which is why they're offering researcher access and keeping it on Why,.
1
1
9
RT @slow_developer: Roman Yampolskiy says we’re building something thousands of times smarter than us -- and we won’t predict its moves. Su….
0
51
0
RT @madmanweb: The AI’s Existential Crisis: An Unexpected Journey with Cursor and Gemini 2.5 Pro .
0
4
0
RT @AllenInstitute: The fireworks in your mind. 🧠✨ This sparkling video shows the neurotransmitter glutamate being released into synapses,….
0
207
0
RT @anilkseth: Do you see circles or rectangles? How culturally dependent is perceptual experience? Inspired by two recent studies, I wrote….
0
25
0
RT @vitrupo: To scale Perplexity, CEO Aravind Srinivas says large models now label, evaluate, and train smaller ones. The system handles m….
0
55
0
RT @dioscuri: Meeting people with <3 ASI timelines who still want to to do a 5 year US PhD 😬.
0
17
0
RT @skdh: some scientists are now hiding AI prompts in papers that instruct potential AI "peer" reviewers to accept the paper 😂 https://t.c….
0
259
0
RT @Sauers_: "This leads to the other major flaw in your thinking: you have no 'taste.'" -Gemini, to Claude (yes, seriously) .
0
30
0
If this is confirmed, scoring 45% - even with reasoning - in Humanity's Last Exam is an incredible feat.
Grok-4 is the new SOTA on AIME 2025, HLE, GPQA and SWE-Bench. 45% on HLE with reasoning is MORE THAN double that of o3.
0
0
2