Whoever gets AGI first owns the future though, any GDP put into manufacturing not essential to that goal is a geopolitical opportunity cost
If a company gets to AGI a month later, why does that matter so much?
We’re not talking super intelligence here, just human level intelligence.
OpenAI was first to ChatGPT yet other companies are still in the game.
1. The first company to get AGI will likely have a multitude of high-leverage problems it would immediately put AGI to task on
2. One of those problems is simply improving itself. Another is securing that company's lead over its competitors (by basically helping every employee at that company do better at their job)
3. The company that reaches AGI for a language-style model will likely do so due to a mix of architectural tricks that can be applied to any general-purpose model, including chip design, tactical intelligence, persuasion, beating the stock market, etc
These things are being commoditized, and we are still at the start of the curve when it comes to hardware, data centers, etc.
Arguing for an all-in civilization/country bet on AGI given this premise, is either foolish or a sign that you are selling "AGI"
AGI is not super intelligence.
Perhaps it's semantics, but I'd say AGI is around the average human intelligence, not extreme outliers like mediocre AI researchers.
GPT-5 required less compute to train than GPT-4.5. Data, RL, architectural improvements, etc. all contribute to the rate of improvement we're seeing now.
Computer science hubris at its finest.
It needs to learn new information, create novel connections, be creative.. We are utterly clueless as to how the brain works and how intelligence is created.
We just took one cell, a neuron, made the simplest possible model of it, made some copies of it and you think it will suddenly spark into life by throwing GPUs at it ?
LLM's can do all those things
Any novel connections are through randomness, hence hallucinations instead of useful connections with background knowledge of involved systems or concepts.
About creativity, see my previous point. If I spit out words that go next to eachother, it won't be creativity. Creativity implies a goal, a purpose, or sometimes by chance, but utilising systematic thinking with understanding of the world.
I feel that many people who deny the current utility and abilities of large-language models will continue to do so far after they've exceeded human intelligence, because the perception that they are fundamentally limited, regardless of whether they actually are or if their criticisms make any sense, is necessary for some load-bearing part of their sanity.
I am denying that they are intelligent, or that by scaling them / upgrading them they will suddenly spring to life and become AGI.
The same arguments are always brought up, often short pithy one-liners without much clarification. It seems silly that despite this argument first emerging when LLM's could barely write functional code, now that LLM's have reached gold-medal performance on the IMO, it is still being made with little interrogation into its potential faults, or clarification on the precise boundary of intelligence LLM's will never be able to cross.
>Proof: I took a convex optimization paper with a clean open problem in it and asked gpt-5-pro to work on it. It proved a better bound than what is in the paper, and I checked the proof it's correct.
https://x.com/SebastienBubeck/status/1958198661139009862
(please excuse the x link)
If you want to hand-wave that away by stating that any company with technology capable of achieving AGI would guard it as the most valuable trade secret in history, then fine. Even if we assume that AGI-capable technology exists in secret somewhere, I've seen no credible explanation from any organization on how they plan to control an AGI and reliably convince it to produce useful work (rather than the AGI just turning into a real-life SHODAN). An uncontrollable AGI would be, at best, functionally useless.
AGI is --- and for the foreseeable future, will continue to be --- science fiction.
The second is a significant open problem (the alignment problem) and I'd wager it is a very real risk which companies need to take more seriously. However, whether it would be feasible to control or direct an AGI towards reliably safe, useful outputs has no bearing on whether reaching AGI is possible via current methods. Current scaling gains and the rate of improvement (see METR's horizons on work an AI model can do reliably on its own) make it fairly plausible, at least more plausible than the plain denial that AGI is possible I see around here with very little evidence.
is the assertion I am referring to; you do a lot of handwaving to assume that AGI is achievable.
There are far more signals that AGI is going to be achieved by OpenAI, Anthropic, DeepMind or X.ai within the next 5-10 years than there were of any other hyped breakthrough in the past 100 years that ultimately never came to fruition. Doesn't mean it's guaranteed to happen, but to ignore the multitude of trends which show no signs of stopping, it seems naive in Anno Domani 2025 to discount it as a likely possibility.
So agi before autonomous tesla? "Just two more years guys I promise", how can people keep falling for these lol
And as the internet deteriorates due to AI slop, finding good training material will become increasingly difficult. It's already happening that incorrect AI generated information is being cited as source for new AI answers.
I'm sure most companies have understood the "AI outputs feeding AI's" incest issue for a while and have many methods to avoiding it. That's why so much has been put into synthetic data pipelines for years.