Reminds me more of The Purge.

I think it’s saying you can’t escape Twitter posts being reposted everywhere.

Like this utterly valueless reply from @vincedmonroy, when we could have just had a quote of Marc Ruffalo’s statement from the Oscars, which other commenters have posted.

Unironically this behaviour is just “pivoting to a run for office as a Republican” vibes nowadays.

Its no longer even ‘weird behaviour’ for a US CEO.

His opinion is actually that AI can use his code no problem, they just have to pay a fee.

The problem is that the big LLM AI companies will just say… ‘Fuck off’, because they don’t like paying for any data, and they also think their models will be advanced enough to write their own libraries soon (if not now, depending how much they believe their own marketing hype).

Pricing is an additional unanswered problem in his new model. As a hypothetical: if 1000 traditional OSS users generate $1000 value in conversion to paid users in his old model - what would an AI license cost? Because one license (eg to Anthropic/Claude) would theoretically be cutting off millions of users, maybe 80%+ of his userbase. Would he ask for millions as a licensing fee?

Whole idea is half-baked IMO, but I am sympathetic to the bullshit situation he finds himself in.

It is a four year contract. OpenAI is hoping they’ll be able to suppress their competitors long enough to regain their lead and firmly established a dominant position in the market.

I’m not too worried though for two reasons. First, I’m confident they’ll eventually be in breech of their memory contracts for being unable to pay - as the whole AI market is a house of cards, and has no real path to profitability beyond hopes and dreams. Banks and angel investors will eventually start asking ‘where are the profits’ and begin pulling out the rug. Second, the chip suppliers began ramping up production (as you suggest) some time back, so the current crazy price increase should only be temporary once they have increased supply output in a year or so. They would have to sign new contracts to get their ‘40% deal’ again, and the memory giants will have much higher price demands for any such deals in future, and I don’t think OpenAI will have the money.

I know, tbh it’s only 50-50 I wash my car - but hey, new year let’s be optimistic.

Wishing you a good one.

This is wrong. The truth is far worse.

Sam Altman / OpenAI recognized that they were losing their LLM market lead to rapid advances by Google Gemini and others, so they took the most anti-competitive step they could.

They determined that the key inputs for AI advances and market leadership now were access to high speed storage and graphics processing - whether directly or via contract to datacenters as-yet unbuilt.

They already had significant contracts and share-trade arrangements with Nvidia - whom is happily gouging the AI Bros for all they can. What everyone in that market needs though, is high speed memory chips for SSDs, RAM, and graphics card memory. So, they secretly negotiated two contracts simultaneously with the two largest memory chip manufacturers to aquire ~40% of the memory market supply.

They have agreed to buy memory chips wholesale, unuseable until they go through further manufacturing to install them intp RAM/SSDs/GFX - but OpenAI has zero facilities or contracts to perform those steps, and as yet has made no public announcement (that I’ve seen) of their actual plans for what to do with the chips they’ve entered contracts to buy.

It is a strategy called ‘market denial’, and we are all paying for it with much higher prices for anything that needs these chips or is tangential to those markets.

• AI bubble will pop when the datacenters that are financed to be completed in the first quarter fail to meet their deadlines, and continue to fail to meet them mid-year. Banks will pull loans, and rapidly, the dominoes will fall. I doubt OpenAI will survive 2026 (fingers crossed), Nvidia may, since theyre actually shipping products and making sales off the bubble - they’ll take a big hit, as will Microsoft, Google, Facebook. It will cause a worldwide recession if not a financial crisis. 75% chance by EOY.

• Trump will die. His health has circled the drain for years, and there’s only so many medical interventions that can be applied. 50% chance.

• I’ll finally wash my car (been putting it off all 2025), 95% chance.

Prudism. That’s the take.

Rather read Oglaf any day than this pearl-clutching nonsense.

Least fucked option.

Reddit: fucked. Twitter: fucked… Meta: fucked since forever. Bluesky: signs of future fuckededness. Mastodon: not what I was after.

Lemmy: hey, these people seem to be my kind of insane.

Internet Copilot One X.

Initial study sample size was 45,000…

Follow-up study (which confirmed initial findings) was the small one.

Fair and reasonable. I don’t use it often and have nearly finished moving entirely to Linux, otherwise I may have done the same (its Windows-only).

So, wait… pay for search, which we need you logged in for and ‘we swear is private bro, honest - but oh hey, if you want a search that’s definitely actually private and we promise it (harder)’, pay them more for a Professional/Ultimate/Team plan to unlock Privacy Pass access … that is uhh, a fresh red flag.

Further, you can’t access Privacy Pass (PP) searches via their standard search engine page… you can only access it by installing and using their closed source browser, or their closed source Android app, or their closed source browser extension… So again its just ‘trust me bro’ but you’re paying them more, and each of the technologies they require you use to access the Privacy Pass can theoretically track all of your data in far more detail than a search engine alone? Mate, from my perspective it’s privacy red flags all the way down.

I mean I get it, how do you have a premium search engine that your users pay for to avoid ads, while also identifying that they’ve paid, while guaranteeing their anonymity/privacy? That’s a tricky thing to solve, but the way they’ve opted to solve is arguably even more suspicious. Open source client code (at least) to validate the server service could not be using the PP tokens to link to a specific user would be a right way.

They’ll deserve it, the shit we’ve put them through.

I feel there’s very little the rest of the world should emulate of modern Russia, but this is one example that I would like to see mimicry of: https://en.wikipedia.org/wiki/Monument_to_the_laboratory_mouse

I didn’t downvote, but probably because they’re a young USA-based search engine that requires login to use - which is usually a huge red flag for privacy, and their privacy of user searches is claimed but has never been verified by any kind of audit - another significant red flag.

Why trust another for-profit Palo-Alto search company with your search data, assist their (potential) tracking by logging in, and pay for it in the process?

I’ve never seen ads and I use the usual free TeraCopy at home. Are there supposed to be ads? (I don’t have a pihole or anything that’d be blocking the ads at the network level)

A Chromecast is not full of ads, but are they all just Google TV now? Is Google TV full of ads? I haven’t used one.

Could be an option to reset your TV, disconnect it from the Internet, and buy & use a Google TV device instead. The streaming devices seem to have far fewer ads and shit than TV manufacturers cram into their devices nowadays.

The hoops we have to jump through to minimise surveillance capitalism… I see elsewhere in the thread you’ve had to use a PiHole to block most of the TVs traffic.

Ask Elon, he tries to come up with a worse one for each child.