Gary Marcus发推怼了一波“AGI近在咫尺”派，拿Astra举例说数学强不等于啥都强，别被带节奏。
Gary Marcus在推文中批评“AGI近在咫尺”群体反复犯“所有认知等同”的逻辑谬误，称今天至少见到六次。他以Astra为例，指出数学出色不等于能避免幻觉或可靠读取PDF。Marcus还怀疑Astra能否在2024年与Miles Brundage的赌约中拿到5/10分。他强调单一领域专长不代表全面型智能，更不意味着AGI或ASI临近。
fail to understand this fallacy at your peril
fail to understand this fallacy at your peril Gary Marcus @GaryMarcus The “AGI-is-near” community keeps committing the same logical fallacy over and over; I have seen it at least half a dozen times today alone. Every time there’s an advance, I see the same error. Here’s how the fallacy works. 1. Someone pretends that all cognition is created equally. (Totally untrue.) 2. Whenever AI achieves success on some form of (fancy) cognition they want you to believe that success on all forms of AI is imminent. I include three prominent examples from today below, each promising that some grand universal can be solved (“science”, “every discipline”, “every problem [people] face in life”), simply because there was an advance in a particular domain (today, math, other times coding, etc). You don’t have to be a cognitive psychologist to realize that this inference just doesn’t follow. We all know that expertise in math doesn’t guarantee genius in all domains. We all know, for example, people who are great at math or physics or programming but struggle with writing or understanding human relationships (and conversely great writers who are weak at math), etc. It’s common to be good at some forms of cognition and not others. Expertise in one domain does not at all guarantee expertise in all or even most domains. That’s precisely *why* people like Gardner and Sternberg developed multidimensional theories of intelligence, why the SAT tests math separately from verbal, etc. Astra is (apparently - we still haven’t seen the methodology) great at math, or at least some forms of math, but that does not mean that it will avoid hallucinations or solve the reliability problems other GenAI systems have. It doesn’t even mean it will be able to read a PDF properly. It doesn’t mean it will be able to follow hard rules either. (Which should terrify you.) In fact, the performance we saw today doesn’t even mean it can write a decent math proof; to the contrary, the mathematician @henryquantum has already given an example where clarity in the proof was lacking. Whatever you do, please don’t get suckered into the “all cognition is alike” fallacy. It’s not all alike. Astra is obviously good at some problems; but that doesn’t mean it will be good at problems that are hard to formalize. It doesn’t mean it will be magic. It doesn’t mean it’s AGI or ASI or any of that. It’s *very* impressive. But I see no reason whatsoever to think it is AGI let alone ASI. If it can score even a 5/10 on my 2024 bet with Miles Brundage I will be surprised. Enough with this fallacy. 🔗 View Quoted Tweet 💬 2 🔄 0 ❤️ 5 👀 1687 📊 2 ⚡