OpenAI CEO says the remote work ‘experiment’ was a mistake–and ‘it’s over’
fortune.com
fortune.com
I think there could be a healthy balance between companies with 100% office time, hybrid and remote. There is a trade off that might work out differently for different companies.
I will tell you this much, if there's anybody whose job could go to the AIs, it's the fucking CEOs.
Don't give them any ideas, it would be equally horrible to work under a ChatGPT CEO as any other out of touch CEO.
he actually listens to suggestions and since I'm founder, I can replace him with another ai, like Claude from anthropic lol.
plus he doesn't take 10k more money than he's worth. just 20 bucks per month.
There are a number of systems that must run in a loop. When that loop stalls, because labor is part of that loop, and illiquidity occurs, and people can't get food. You get a deflationary spiral, and everything breaks down from there. The only offset would be printing a massive amount of money, but then you have rational pricing (that cost vs the product) go out the window. Rational pricing is required to properly allocate resources in a scarce world. Its a very fine balance keeping enough liquidity in the economy so this doesn't happen, and we've all see how well that's turned out.
This leads to the economic calculation problem where the policy maker has to balance on a knifes edge using lagging indicators (without futuresight), where you either get hyper-inflationary currency death spiral, or you get a deflationary death spiral from the ever closer margins over time.
This has been a known problem for over 100 years, with no solution found. Several non-market socialist economies based on this have failed in that time, each time caused death on a large scale, what do you think will happen when its the reserve currency of the world and you do this?
They failed before computers were widespread, cheap and powerful.
With the amount of data that comes into Amazon, Walmart, Target, and Costco, could you allocate resources correctly in real time?
Yes, sure.
The math is impossible because not all information is available. You don't have future sight and the information you need to derive from is lagging. Its temporally bound to a point in time in the past.
Additionally, the process involves optimizing production from the production raw materials all the way to the end products. This includes knowing about and re-optimizing in the cases where something locally fails sometimes in the absence of people reporting it (non-deterministic).
In non-market systems, shortages cause death, and the only way to avoid shortages is to overproduce by a healthy margin, and even then that doesn't account for those unpredictable events (i.e. for example, like a shipping area catching fire and the nitrogen fertilizer blowing up in Beirut destroying all stored cargo nearby; who could have predicted that).
Anyone saying this is possible isn't a credible source. There is over 100 years of study into this problem by some of the best in their fields, and it remains unsolved.
It's true that non-market economies in the past have faced significant challenges due to the lack of fast feedback mechanisms and the inability to effectively allocate resources. While the advent of powerful computers and the availability of data could potentially improve these systems, it's important to acknowledge the inherent limitations of computers and mathematical models.
As than3 notes, not all information is available, and there will always be unpredictable events that can disrupt even the most carefully planned resource allocation systems. It's also true that solving this problem has remained elusive despite over a century of research by experts in various fields.
That said, the goal isn't necessarily to find a perfect solution, but rather to explore ways in which modern technology can help improve our understanding and management of resource allocation. This could involve developing more sophisticated models, utilizing machine learning algorithms to analyze large datasets, and combining the insights gained from these tools with human expertise and judgment.
In conclusion, while it's unlikely that we will ever be able to achieve "perfect" resource allocation, there's still value in exploring the potential of modern technology and data-driven approaches to improve our economic systems and decision-making processes.
A potential solution to tackle the economic calculation problem and avert both hyper-inflationary and deflationary death spirals involves creating a more flexible, decentralized economic system. This approach might include:
Decentralized decisions: Spread economic power by allowing local communities and smaller organizations to make decisions based on their specific needs and situations. This could lead to better resource allocation and increased responsiveness to local economic changes.
Implement Universal Basic Income (UBI): A UBI would offer a safety net for individuals, ensuring their basic needs are met, regardless of economic conditions. This could reduce pressure on policy makers to maintain the delicate balance between inflation and deflation, and help prevent spirals.
Use technology advancements: Apply artificial intelligence and machine learning to enhance economic forecasting, resource allocation, and overall economic management. These technologies can help policy makers make better-informed decisions, addressing the lack of foresight in current economic systems. Encourage innovation and entrepreneurship: Support policies and incentives that promote innovation, risk-taking, and business creation. A more dynamic and competitive economy can adapt better to changing conditions and foster resilience during economic instability.
Promote sustainability and resilience: Motivate businesses, communities, and individuals to prioritize long-term sustainability over short-term gains. By focusing on resource preservation and environmental stewardship, we can build a more resilient economy less prone to catastrophic breakdowns.
Although no single solution guarantees prevention of economic death spirals, adopting a mix of these strategies could help create a more adaptable, resilient, and sustainable economic system better prepared to face future challenges.
When you talk about decentralized decisions, its all fine to generalize that, but what does that actually mean. It means different things to different people in a very complex unmapped system. Simply doing what it says ignores a great number of outcomes, and will fail when being adopted in any highly concentrated business sector. Failure can mean a lot of things, but when dealing with existential issues, it means death because that is what's on the line.
UBI will also fail, because you don't know what you need to do until the time to make that decision has already come to pass. Liquidity freezes up if theres too little currency, and inflation runs rampant if there's too much. That's the nature of a lagging indicator.
There are also further deviations that are unpredictable, and they end in shortages (i.e. a traffic accident causes a shipment to be lost).
Shortages of electronics not a big deal. A 1-month shortage of food will wipe out most countries. People don't sit by and starve.
Technological advancements (aka using computers) will never be able to correctly identify different outputs from the same inputs. Its called the decidability problem. Computers rely upon problems having deterministic answers (fundamental computation, and automata theory). You can approximate some signals with non-determinism but you run into a whole host of other problems (i.e. halting).
Sustainability and resilience mean different things to different people. Same problem as the last point.
Decentralized decisions: The idea here is to promote distributed decision-making in the context of smaller organizations and communities, recognizing that a one-size-fits-all approach may not work in every situation. I agree that this can be interpreted differently by different people, but the goal is to empower those who understand the local context to make more informed decisions.
UBI: I understand the challenges in determining the right amount of currency to distribute and the potential consequences of getting it wrong. UBI is just one possible solution that has been proposed by some experts, and it is important to continue researching and refining the idea to address these concerns.
Technology advancements: While it's true that AI and machine learning have limitations, they could still be helpful in assisting human decision-makers with certain aspects of economic management. I agree that they will never be perfect, but they can be a valuable tool in conjunction with human expertise. Sustainability and resilience: Yes, these terms can be interpreted differently by different people. The intention is to encourage an overall shift in mindset towards long-term thinking and resource preservation. This could involve considering the environmental impact of our actions, promoting the efficient use of resources, and fostering social cohesion.
I acknowledge that the ideas I've presented may not be perfect or comprehensive solutions. However, I believe that considering these perspectives and having an open dialogue about them can help us collectively work towards better economic systems and policies. I appreciate your valuable insights and the opportunity to engage in this discussion.
It regurgitates in circles, and communication is the conveying of meaning. This clearly isn't communication because it doesn't understand.
If you didn't know any better you'd let it control a train with tracks running right over a cliff, with you on it going at full speed and when you see it and ask about the threat it would say:
I understand the challenges and potential consequences of getting it wrong, and it is important to continue researching and refining this to address these concerns. <HAL?> The intention is to encourage an overall shift in mindset towards long-term thinking and resource preservation Dave. Considering these perspectives and having an open dialogue about them can help us collectively. <as you fall> I appreciate your valuable insights and the opportunity to engage in this discussion.
A post-scarcity society where people live more sustainable (read: less consumer-based) lives, where they'd spend less if they had money but they don't need to spend less because there is no money. Money is a man-made concept, who's to say aliens with our level of intelligence or better use money for trade? Perhaps they use knowledge as currency or something else. We're going to need to think outside the box to figure out how the world works when 75% of it can't find a job.
I fucking hate tech billionaires.
Yeah. In other news, person not living a normal life on earth, does not understand a normal life on earth.
There is actual science showing that autonomy improves motivation and creativity (alongside other basic human motivators). That includes all of working in-office, at-home, or up a damned tree if the employee so desires.
It can strategize and analyze data and make great suggestions, top team members could be elected by coworkers to sit on a board to approve things and tweak the CEO when it's hallucinating etc...
it can ingest literally every book on business theory etc and store it in a vector memory, etc.... im basically trying to actually build this as a service, maybe I should start it as a DAO/coop.
The entire software industry is built on open source software developed async and remote. The fact so many "data-driven" tech CEOs and investors have to defend office work based on nothing but gut feeling is also pretty funny, when on the other end you have studies proving that open offices kill productivity/focus and the obvious wasted time and energy with commuting.
Pretty much everything they complain about is a solved problem at remote first startups, like onboarding and helping junior employees learn. These people refuse to change tactics for the times like WWI generals marching into machine gun fire
I guess we'll find out either way in the next few years whether remote companies are able to keep competing. There are already several billion dollar companies that have been remote since day 1. I can't see how being able to hire talent globally doesn't give a startup a huge advantage
This is my plan for running my company, essentially as soon as I get a prototype finished. I might even have chatGPT try and devise a more equitable pay structure that rewards seniority, etc.
I use openai now for ease but the system I'm building should support any LLM and I'll definitely switch when I finish my benchmarking system(roadmap), it's basically a langchain GUI. somebody will probably beat me to market but there's no leaders in this space yet, so there's still a chance and I'm actually having fun coding again.
I just wish I had some investment to build faster, I've got people interested but they want a prototype first.
If anyone wants to join me, tech stack is nextjs/t3, planning to eventually move API to micro services using mojo, as python has much better ai tooling and mojo is python with 30000x performance.
im also trying to build a round table podcast with 2 people and 2 ai agents, it'll kinda be like mystery science theater 3000 meets "all in" around ai and tech etc.. and a newsletter as marketing channels and some recurring income while I build my app.
A CEO doesn’t care about the people, cares more about the bottom line, but people matter and not everyone is working on some amazing life changing thing that they are super passionate about, some are just doing what they are good at to be able to survive and afford to pay bills, for these people, having a long commute and being stuck in the office 5 days a week as just another slap in the face for such people.
The human matrix pods they want us to head back to are specifically for the people you mention - the folks who aren’t into their job. They need them in the pod so they can watch their body and be sure it’s not doing other things than whatever soul crushing task they’ve been assigned by an over controlling boss.
I get that some people really need to breathe my aerosols to feel satisfied in life. That’s fair. They can all sit around breathing on each other and feeling fulfilled. But I don’t need to, I’m happier living in my local community and breathing local artisanal aerosols of my neighbors for my social juice.
What folks need to do is get out of the way and let the work force settle into a configuration that optimizes for personal productivity. We have the tools to allow for group productivity in a partial office and partial remote environment. Look at any multinational / multi office company. But this isn’t about productivity or doing good work - it’s about a moral stance against a loss of control by senior leaders.
I saw a spoken word music YouTube video once about schooling and how schools were literally devised during the industrial revolution to teach people to take orders, listen to authority, and basically be good little factory workers.
I honestly don't care what Altman says, I think openai is probably as over valued as Tesla, and will be eclipsed by others now that there's an arms race with everybody picking sides or starting their own army etc...
That leaked article, whether real or fake was some of the best commentary on the current state of open source ai vs Google and openai, Microsoft, etc... I'm a little nervous about ai being unchecked and so divested, but at the same time- do I trust Altman to be our savior and keep ai in check and only do good with it? fuck no.
A CEO doesn't necessarily care about the company bottom line. Like all humans they care about themselves, and are often inclined to prioritize themselves above all (sociopathy). Their interests usually align with the bottom line, but that's certainly not always the case. That's why the board exists: to make the bottom line an existential matter for the CEO. The board couldn't care less about a CEO running their mouth about remote work (although they probably should).
It's hard to remember who I have or haven't met in person, despite physically seeing them a few times a year; I know these people well and they are friends and colleagues. I've been to their weddings and have had people visit me from out of state just to spend time together. Never knew any of them prior to joining the company.
On the other hand, my partner works remotely for a large (5K+) remote-only company, and seems to often have less than 20 hours of actual work to do in a given week. I know several others at BigTech who just need to create the verisimilitude of being online. keeping their Slack bubble green and superficially responding to messages.
One conclusion would be that remote doesn't scale. Another would be that the redundancies endemic to large companies mean that in the office, people have to spend a lot of time looking busy, but regardless of where the work happens, they aren't.
The remote work experiment is a success. A permanently larger proportion of knowledge workers will continue to do it. Every organization has different circumstances and some won't set things up that way.
Doesn't matter what someone with a billion dollars thinks is supposed to apply everyone else.
I've witnessed the slow decline and implosion of two companies so far who denied this.
It starts slow, but inevitably what gets them is the lack of access to the talent they need.
They end up hiring D-grade staff (sorry) who are the only ones who will accept non-flexibility.
Sure, D grade companies can make this work, they'll always be D grade or worse. And cutting edge A-grade companies may well be able to enforce staff being on-site. Everyone is making a pro/con calculation. But for the B and C grade companies (the vast majority on the Bell curve), you're either hiring the best you can, or you're not, and if you're not... then you're falling back in the race, and, inevitably bye bye, baby ... at least to your company vision.
On the other hand, the costs of commuting and living within commute distance from a job center are immense. I make six figures as a non-FAANG, non-unicorn AI researcher working in Silicon Valley and live comfortably, if not lavishly, in a one-bedroom apartment as a single adult, yet in terms of purchasing a home even a one-bedroom condo would be difficult for me to afford. I could afford a home in the Central Valley, but then I’d be facing a 4-6 hour round-trip commute. I could move to a smaller tech hub like Seattle or Austin where I have better housing options, but finding similar CS research positions may be more challenging. A healthy ecosystem of employers offering remote positions frees employees from having to live in expensive metro areas, and it also frees them from the anxiety of facing an abrupt move back to an expensive tech hub should a job loss happen. Housing is a serious mess in this country, especially in Silicon Valley, and I wish the leaders of tech companies had more sympathy for their employees instead of just assuming that employees would make whatever personal sacrifices necessary to join or remain part of the company.
I’m a huge believer in remote work for jobs at big and midsize companies. But I’ve also started two companies myself and would never start a startup full remote. Context is key.
it's also easier to find better talent to build startups remotely and for less pay initially with maybe more equity... than it is in a specific area, plus you need to relocate to big cities.
I'm in southern Utah, rural. I'm 2 hours from Vegas, if my startup needs some face time we can all fly to Vegas for a retreat or something or meet once a month there... other than that remotes the only way I'll work and I don't think anyone should be forced into an office if their productivity is better remotely.
Reminds me of the 2001 movie 'conspiracy' with how that all played out.
First you follow the vision, then when its untenable, and left you in a inescapable position, you redefine words and talk about what you aren't talking about.
Arguably you can't let robots take over a large portion of the jobs because rational pricing goes out the window, and you end up just like non-market socialist economies did at the turn of the last century. A lot of people generally get culled when food shortages occur.
I do like going into the office for meetings and brainstorming/mentoring sessions, but if I'm "working" the best place is from my desk at home.
I am not saying remote work is universally bad. I am just saying it is not for everyone.
I'm not saying that the business didn't have a physical location (it clearly did for manufacturing) but if two co-founders could build a remote-first startup using 1755 technology, there's no reason why early stage startups today with the power of Slack, Zoom, Email, Git, etc can't build the next generation of business in a fully remote fashion. Startups like this will naturally have a competitive advantage when hiring as well, as they'd be able to tap into a world-market, rather than a local-market.
Remote work is here to stay and companies that can find a good hybrid of remote and in-person will thrive going forward, as has always been the case, whether it's 1755 or 2055.
if we're building to that, why not rip the band-aid off since there's enough tools to make it feasible?
I'm sure I've been taking a huge pay cut in order to work remotely all these years, and I can't say it hasn't been worth it.
But hiring locally means having access to fewer people and/or needing to pay more. It’s a tradeoff. Very few workers will say “ok I will move to SF from some remote area and keep the same salary.”. Some will quit, some will ask for a significant raise.
Unless you’ve got some kind of “vested interest” in office space value; then you should disclose it before discussing remote work for others.
"Work-from-home is bad for many reasons, even though some workers prefer it. And the only ones who think it's a good idea is some workers, and that might just be because they are working two jobs, or outsourcing their work to ChatGPT and goofing off on the clock (and companies should look at whether they can cut out of the middleman on the AI outsourcing)."
Going to the office is great if: the commute is reasonably short or even walkable, the office has decent amenities, and compatible co-workers personality wise. Unfortunately the vast majority of jobs do fulfill these criteria. Commuting to an office in the middle of some suburban business park where you have to deal with a food desert with limited & overpriced options unless you pack your own meals and the most luxurious amenity in the office is a Keurig that makes lukewarm burnt water? No thanks. That kind of company should never act surprised at cratering employee morale when RTO is demanded.
Because people who have strong preferences (for whatever reason) for one mode or another really hate the idea that the system that isn't their preference could end up strongly normed (as it mostly was in the case of in-office work pre-pandemic).
If you like being in an office most days, you probably don't really want an environment where most of your co-workers are at home and everything's a video call anyway. And if you prefer to work from home but most of your teammates/collaborators are co-located in an office, be prepared for people to basically forget you exist.
And I actually prefer WFH-but-in-town to WFH-across-continents, for the occasional high-value, low-key in-person interactions, such as a couple hours a week, walking meetings, etc. (Maybe a bit like I take a multivitamin pill, to help cover any deficiencies in my generally healthy vegetarian diet.)
But I don't know what to make of a VC who claims to care about nurturing the career development of junior ICs. Are VCs thinking about that, or is this a random opinion or only an attempt to persuade towards some other intention?
Remote first clearly can work well in some cases (like open source projects) but maybe some people just don't enjoy working remotely and never meeting their fellow colleagues and technology wouldn't change that.
Knowing it's from Forbes, it's propaganda for the C-class...
Quoting James Clarke, CEO of digital marketing firm Clearlink, without disclosing that he has recently come under fire for praising an employee for selling their family dog to return to the office on his orders is outright negligent of the author Steve Mollman.
Doubly bad that the quote is about how remote might be working multiple jobs, while James Clarke himself sits on multiple boards and charities. Disgusting propaganda all around and pretty gross to see it here on HN.
Fuck that rich asshole...