Hacker Newsnew | past | comments | ask | show | jobs | submit | Retric's commentslogin

Depreciating assets is one of several possibilities.

That argument supports any levels of losses, however I also think it’s rather misleading.

Growth means some inefficiencies, but their expenses are largely around commodities like electricity and data centers not a sudden army of salespeople. They also got 150M 11 years ago and 1 billion 7 year ago, they where quite large in 2022.

Basically you don’t get better at writing checks to your local utility which limits how much they can control costs.


> They also got 150M 11 years ago and 1 billion 7 year ago, they where quite large in 2022.

I had to look that up, you're talking about investment there, not earned revenue.

The 150M was their initial funding (actually 130M I think https://www.clay.com/dossier/openai-funding)

The 1B was from Microsoft in 2019: https://openai.com/index/microsoft-invests-in-and-partners-w...

In 2022 they only had 335 employees (according to various internet searches but I can't find an original source for that number.) I can't find credible numbers for revenue from the GPT-3 API, which did have some usage - GitHub Copilot started charging a subscription fee on June 21, 2022 - https://github.blog/changelog/2022-06-21-github-copilot-is-n... - and that was running on the OpenAI Codex model so presumably OpenAI had some revenue from that.


We don’t disagree with the underlying facts.

That said, in many ways 335 employees is the midpoint between 3 employees and 30,000 employees. The CEO can’t keep track of everyone’s names and what they’re doing, you need layers of management, HR, etc. It’s not really a simple exponential function but 335 to 336 is way more automated than going from 3 to 4.


Fraud seems like a more direct issue.

Someone else isn’t a 1:1 replacement, when people refuse to work for you you’re stuck offering higher wages and or taking worse employees.

How well a job is compensated on average very much depends on how willing and able the average person is to do it.


Do you have specific and objective examples of things people got right?

From my perspective there’s very slow but very real progress happening in the AI space. I see people making wild predictions in both directions, but in terms of actual unsupervised utility there’s definitely progress abet wildly slower than most hype.


This was their prediction for 2026:

"The bet of using AI to speed up AI research is starting to pay off.

OpenBrain continues to deploy the iteratively improving Agent-1 internally for AI R&D. Overall, they are making algorithmic progress 50% faster than they would without AI assistants—and more importantly, faster than their competitors. The AI R&D progress multiplier: what do we mean by 50% faster algorithmic progress?

Several competing publicly released AIs now match or exceed Agent-0, including an open-weights model. OpenBrain responds by releasing Agent-1, which is more capable and reliable.28

People naturally try to compare Agent-1 to humans, but it has a very different skill profile. It knows more facts than any human, knows practically every programming language, and can solve well-specified coding problems extremely quickly. On the other hand, Agent-1 is bad at even simple long-horizon tasks, like beating video games it hasn’t played before. Still, the common workday is eight hours, and a day’s work can usually be separated into smaller chunks; you could think of Agent-1 as a scatterbrained employee who thrives under careful management.29 Savvy people find ways to automate routine parts of their jobs.30

OpenBrain’s executives turn consideration to an implication of automating AI R&D: security has become more important. In early 2025, the worst-case scenario was leaked algorithmic secrets; now, if China steals Agent-1’s weights, they could increase their research speed by nearly 50%.31 OpenBrain’s security level is typical of a fast-growing ~3,000 person tech company, secure only against low-priority attacks from capable cyber groups (RAND’s SL2).32 They are working hard to protect their weights and secrets from insider threats and top cybercrime syndicates (SL3),33 but defense against nation states (SL4&5) is barely on the horizon."

https://ai-2027.com/

That's precisely where we are.

This is eerie. It's like a time traveler. The only delta is Anthropic is in the role of OpenAI.


So you think Anthropic is using internal AI assistants to pull away from competitors and the leapfrogging we've seen over the last several years is now done?

That seems to me to be the most concrete and least obvious prediction in the quoted text.

I don't think that's happening. If that were generally accepted as true I would expect OpenAI to be unable to successfully IPO.


Anthropic would be pulling away if it wasn't so far back because of stupid, arogant philosophy. Instead AI assistants just helps Anthropic stay relevant provided it's hyped sufficiently.

We wait for gemini 3.5 pro and cope?

That's because one of the author's of AI 2027, Daniel Kokotajlo, an ex-OpenAI researcher, was the most prescient predictor of our modern situation from 2021:

https://asteriskmag.substack.com/p/before-he-wrote-ai-2027-h...

https://www.lesswrong.com/posts/6Xgy6CAf2jqHhynHL/what-2026-...

Anyone who wants to dismiss the LessWrong / X-Risk / "doomers" should link their accurate predictions from 2021.


And one was Scott Alexander, who correctly predicted that GPT-2 would scale to general intelligence: https://slatestarcodex.com/2019/02/19/gpt-2-as-step-toward-g...

The idea that progress is “slow” in the AI space is absurd. These are some of the fastest growing products and companies of all time. The models are still improving a surprising amount.

It's not that absolute progress is slow, it's extremely slow compared to the predictions. It might be fast in absolute terms, but the "50% of coders will be obsolete by 2023" has been renewed every six months, and it's becoming increasingly clear that there's a real chance it might not ever happen.

„Coders being obsolete” is not a measure of AI capabilities. I see coders being more busy than ever before. I see people without coding knowledge getting more behind. The gap is widening, not shrinking.

I see people without coding ability catching up to learned coders. AI is a huge force multiplier for people who don't have hands on, detailed technical knowledge, AI can increasingly handle that and just needs a human to steer it more broadly.

It's huge force multiplier for people who have hands on, detailed technical knowledge as well.

3 * N < 42 * N

42 - 3 < N * (42 - 3)

It helps to know layer you're working on.

People seem to make mistake of thinking how good LLMs are around tasks that they are familiar with and extrapolating it to whole population.

It's good mental exercise to think about how little you can do compared to expert on tasks you never thought of working.

Ie. if you're programmer or know something about finance, don't think how much it enables you to do better coding or investing, think instead how much it doesn't enable you to work on something you don't know like maybe molecular biology or visual special effects – it's all there but it's much better multiplier for people who do know their shit.

Knowing layer you're working on helps a lot, it gets multiplied.

Knowing programming is becoming more fundamental skill than ever before as it lies at the foundation of almost everything else.


I suspected you felt that way even though it hasn’t been my personal experience.

I’ve heard people say older models can’t do X, when I used that way etc. I suspect people are applying their own learning curve as part of their assessment of progress, you get better at writing prompts and it feels like the model improved.

Which is why I’m saying we need some objective metrics to judge predictions of actual capacity.


There are objective metrics, that’s what benchmarks are.

Companies game benchmarks, but sure they are better than nothing.

I mean you wanted something objective, and they are. I don’t know why you’re being dismissive of them, they’re a huge element of what drives model development forward.

These companies aren’t just making stuff up, they really do want to improve the models, and the models really are improving.


> I don’t know why you’re being dismissive of them

I’m aware of multiple cases of benchmark cheating/“optimization”. So, taking benchmarks a face value seems laughable.


Depends on what you mean by wages. Floyd Mayweather Jr.‘s career earnings are over 1 billion. Celebrities can get extremely high compensation packages especially when you adjust for inflation without an ownership stake in anything. Finance can be similarly well compensated at the very top.

Often very high compensation packages happen to include shares, but that’s just the form of compensation not an inherent requirement. Of course all paths to a billion dollars are so unlikely it’s really not a reasonable target unless you have already passed most significant hurdles.


The alt right playbook isn’t identical to other political movements because it’s facing different pressures.

Right now it’s dealing with second generation propaganda where much of the leadership believes the narrative rather than the underlying justification for that narrative. This is mitigating by the older generation retaining a great deal of power, but it creates some IMO really interesting dynamics.


> The alt right playbook isn’t identical to other political movements

Straw man. Nobody claimed this. Just that the factor identified, politics by attention economics as a result of social media, is not unique to the alt right.


> this is the basis for all non-retail politics.

My point was there’s an internal disconnect inside the alt right movement which makes this play out in very distinct ways. Dig into say China’s political to social media connections for some wildly different dynamics.

> is not unique to the alt right

Sure, that I can agree with but it’s a long way from your earlier blanket statement.


> Dig into say China’s political to social media connections

China regulates social media in a way we don’t. The fundamental dynamic doesn’t apply there.


Even the US regulates social media. UK, Germany, South Africa, and Brazil are all interesting because of just how different yet similar a role social media plays in politics.

First past the post vs representative representation create some really interesting points of divergence.


> Even the US regulates social media

Federally? Not really.

> UK, Germany, South Africa, and Brazil are all interesting because of just how different yet similar a role social media plays in politics

Could you expand on this?


> Federally? Not really.

TikTok is just one of many examples where the federal government has played a significant role. I mean you can debate about how relevant terrorism, CSAM, etc are here, but lots of debatably minor changes still add up.

> Could you expand on this?

It’s a lot to try and summarize in a comment, but just as an example. UK elections can take place early when a coalition breaks up this places a lot more power in the hands of voters and thus social media mid cycle. In the US passing unpopular legislation early means it’s less likely to be remembered next election cycle.


Somewhere between just going out to vote and revolution sits moving to an area where your vote counts.

I’ve not quite reached that threshold, but I avoid moving to DC due to the lack of voting rights.


He spent that much time and you still misunderstood the direct message and missed the subtext.

The lie is coding is solved, the proof is they had an outstanding coding issue they were working on for over a year while saying coding is solved. There’s a great number of other issues with their own software that disprove their premise, but you only need one counter example to disprove something.

And because you missed it, the subtext was they want you to use loops not because they work but because they burn lots of tokens thus making them more money.


One unresolved bug does not disprove anything.

And saying coding is solved doesn't prove anything either.

A major outstanding software bug for a year proves they have not “solved coding,” which was their claim.

I didn’t chose the words they used, but I can hold them to those words.


The point is if "coding is solved" was true, there would not be any unresolved bugs.

Customers abandon companies that fail.

Remember how they used Brando to water plants and it kills them? Eventually mistakes break critical systems and you fail.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: