Users dismiss as a huge assumption the claim that OpenAI and Anthropic are heading toward the same destination in AI model development, likening it to comparing unrelated geniuses like Von Neumann and Einstein.
Based on 1 visible X reactions from 5 accounts; directional sample.
Ask a question below.
Published answers will appear here.
@TheAIShrink @danshipper Same destination is a huge assumption. Kind of like comparing Von-Neuman and Einstein. Both smart, in different ways!
It just feels all accidental that Frontier Labs chooses a development strategy based on an intuition of how to move forward and changes its trajectories based on discovering failure models. OpenAI discovered reasoning models before anyone else; they abandoned big models (i.e., GPT -4).5) to pursue a bootstrapping approach driven by improved reasoning. Meanwhile, Anthropic continued to stick with its original agenda. Google pursued multi-modality first. These models continue to become more capable despite having divergent development paths.
@danshipper Yes, that seems to be the tell that they may have abandoned that approach in favor of evolving reasoning models like o1. Meanwhile, you didn't see Anthropic change their approach when reasoning models came to the scene. So now we have frontier models that evolved differently.
Meta and xAI push the narrative that industrialists win out over scientists. IMHO, I don't like the taste. But our preferences for certain narratives don't influence what happens. https://twitter.com/47fucb4r8c69323/status/2075435908804079909
@danshipper Did a divergence occur when OpenAI sought the reasoning model route, and Anthropic stayed with tweaking LLM models? Different paths towards greater intelligence?
Conway's Law applied to AI models. https://twitter.com/IntuitMachine/status/2075578505803473089
Related: https://x.com/IntuitMachine/status/2075578505803473089
https://x.com/IntuitMachine/status/2075578505803473089
Users dismiss as a huge assumption the claim that OpenAI and Anthropic are heading toward the same destination in AI model development, likening it to comparing unrelated geniuses like Von Neumann and Einstein.
Based on 1 visible X reactions from 5 accounts; directional sample.
Ask a question below.
Published answers will appear here.
Conway's Law applied to AI models. https://twitter.com/IntuitMachine/status/2075578505803473089