6 ms·
Anthropic Needs to have the most intelligent and scary agents, so if OpenAI does something bad they need to prove their models can do even worse. Without that t
by prymitive 2mo ago
Anthropic Needs to have the most intelligent and scary agents, so if OpenAI does something bad they need to prove their models can do even worse. Without that the whole valuation collapses. There’ll be more “our model outhacks others” for the next hype cycle.
- Schlagbohrer 2mo agoI find the era of "apocalyptic alarmism as marketing" super strange and maybe not good either. How long until one of these firms intentionally "forgets" to airgap the model during a cybersecurity test, just to get a good headline about how powerful theirs is? Perverse incentives all around.