4 ms·
AI in 2025: Predictions by an OpenAI Insider and ex-DeepMind Research Engineer
- crop_rotation 4y agoWow his predictions are even more scary than I thought. The headline "I predict that by the end of 2025 neural nets will: - have human-level situational awareness (understand that they're NNs, how their actions interface with the world, etc) - beat any human at writing down effective multi-step real-world plans - do better than most peer reviewers"
- ftxbro 4y agoHis one specifically about peer reviewers: I'm positive GPT-4 is already better than the median especially if you let it read some of the cited articles. Also I'm not saying this as if I have beef with some who reviewed my papers, I'm saying it about me too if I phone one in. Like, LGTM, or didn't cite x, or didn't label axis or whatever. For anyone who doesn't know, peer reviewers don't check or replicate the substance of the work, they check if there is anything that is obviously wrong with it on a superficial scan of the manuscript and assuming good faith of the authors.
- wsgeorge 4y agoI found some of the replies interesting [0]. In the list of social issues caused by AI advancement, I'm not sure how much focus is given to the potential backlash against the companies and individuals leading us here. [0] https://twitter.com/UsualTomo/status/1640741657162444801 https://twitter.com/UsualTomo/status/1640741657162444801