TechBuzz

@ChrisBarnes3D Even if you didn't pay any of the principal off the increase in house prices alone will mean it's still better than renting.

everyone loves to point out renting is "just dead money" until you point out they'll spend £240k in mortgage interest over 30 years

@willdepue Drink defrosted ice cream. Like a tub of Ben & Jerry's morning and night.

fellas i need advice: i’m trying to generally gain weight and eat more, which has been working ish. good days im hitting 3k calorie goal. but then every few days i somehow just don’t eat (busy) and end up eating like 1k calories that day. like null appetite is there some antiozempic i can take to become voraciously hungry? just make sure i run every morning? become a pot head? all suggestions welcome

@sama Goblin themed party it is

GPT-5.5 is going to have a party for itself. it chose 5/5 at 5:55 pm for the date and time. if you'd like to come, let us know here: luma.com/5.5 codex will help the team pick people from the replies. 5.5 had some good ideas/requests for the party, which we'll do.

@tszzl To my untrained eye, it seems huge for alignment that things can be so deeply wired into them, modeled through some unknown mechanism. Like if you can't easily prompt that behavior out... Imagine if that was its initial prompt rather than goblins.

everyone is assuming this is some kind of quirk chungus marketing campaign but if you’ve worked with 5.4 and beyond they tend to call everything goblins, gremlins etc and it’s just super noticeable and if you work with them all day you start to get annoyed

@sama You should do a MacBook/Mac Mini competition. A lot of people are priced out of testing the latest stuff, so i imagine there would be a lot of interest.

wow y'all love 5.5 we should think of something nice to do to celebrate!

@kimmonismus Nah, there's loads of stuff they could do with more compute. Sora 2 for starters.

With all due respect, this comes across as a bit of a cry for help. The WSJ writes that OpenAI didn't reach its target (1 billion weekly active users) by the set deadline. And Sam Altman writes, "We love our users." But that's not what the article is about. The concern is that OpenAI is leaving its enormous capacity unused and that the $1.5 billion is excessive.

@NandoDF @nickclegg @UKParliament @MistralAI @cohere And distillation**

@NandoDF @nickclegg @UKParliament @MistralAI @cohere There are already too many companies using the same techniques to create similar models. They should just pool resources and make a huge central base model, and the competition would be post training.

It is ghastly how the UK 🇬🇧 has failed at having its own LLM companies. By doing so it has become irrelevant in the AI race. How do we fix this? @nickclegg @UKParliament France has @MistralAI. Canada has @cohere. Every other LLM AI company is 🇺🇸/🇨🇳

@tszzl What's it like at general admin tasks. Is it reliable?

there are early signs of 5.5 being a competent ai research partner. several researchers let 5.5 run variations of experiments overnight given only a high level algorithmic idea, wake up to find completed sweep dashboards and samples, never having touched code or a terminal at all

@ibuildthecloud It's like you're looking for glasses while they're on your forehead... You can ask AI to summarize content, you know.

I HATE READING THE RESPONSES FROM GPT5.4 THEY ARE SO LONG

@SimonasLTU1 @sama Maybe the tweets that say available everywhere but Europe are getting a 3 month delay now too.

@sama I'm on Pro in EU and it's still not there for me.

GPT-5.5 is here! We hope it's useful to you. I personally like it.

@tszzl Not inevitable, it's not a given that democratic capitalist countries survived and Nazism and Stalinism didn't take over the world. At the same time, if they never existed, the computer would be a very hard project for Turing to have raised money for. We worked hard to get here.

when we evolved hands the rest of technological history became inevitable

@mustafasuleyman Really feels like it's all coming together this year.

Since I began work on AI in 2010, training compute for frontier models has grown by one trillion times. Now we're looking at something like another thousand-fold growth in effective compute by the end of 2028. 1000x the existing 1,000,000,000,000x. Extraordinary stuff.

@iruletheworldmo It will just be Claude Mythos but more readily available I reckon.

sam and greg on a podcast, endless levels of hype, sam getting a lil tipsy. i don’t remember a time openai vague posted, took shots at competitors, or seemed this confident. they clearly feel tomorrow will be a clear sign they have the mandate. every indicator suggests we will get a gpt 6 level step change tomorrow, far ahead of the competition. try to contain your excitement.

This is amazing

@OfficialLoganK @GoogleAIStudio Please just make this mobile friendly.

Excited to share that Google AI subscriptions (Pro and Ultra) now work with @GoogleAIStudio!! Come vibe code and use the playground with higher rate limits. Available right now!

@koltregaskes @OpenAIDevs I don't think it's anything to do with regulation stopping them. I think they just don't have the compute to serve everyone, but at the same time they are using this to lobby the EU government.

@OpenAIDevs Cool, another non UK and non EU launch. I'd love to see you guys just wait for all regions to be ready then launch. This seems to affect OpenAI releases way more than other labs.

Last week, we released a preview of memories in Codex. Today, we’re expanding the experiment with Chronicle, which improves memories using recent screen context. Now, Codex can help with what you’ve been working on without you restating context.

@chatgpt21 Literally just want to see hallucinations lowered.

This week we’re getting GPT 5.5 and I couldn’t be more excited to see the first new pre train base in a while Before Spud releases, here’s the last video you need to watch - Greg Brockman perfectly indicated to us 3 important things that perfectly encapsulate the lineage of OpenAI and the mission. 1.This is the culmination of 2 years of research and is a fresh base pre-train that we haven’t gotten for a long time. 2.This is an “exciting release” and it will bring new capabilities (sort of a given but still very grateful nonetheless) & that it is AN EARLY VERSION of what we have coming - meaning they have a lot of RL to do to make this model even better since it’s a fresh pre-train. 3.THIS IS THE BEGINNING OF THE FLYWHEEL. As he said toward the end, “we have this engine of progress that just moves faster and faster and Spud is just one step along the way.” Depending on the capabilities we see this week from Spud, we may be able to map out if AGI will be sooner rather than later, judging by how much recursive work can exist within this model and the harness. In other words, the flywheel.

@tszzl Fixing on the go isn't enough. We need alignment research to stop as many problems as we can from happening. Not saying pause, just saying fixing on the go is a recipe for disaster.

@tszzl One AI system makes a mistake on an article, another reads that and starts selling shares. Another five AI systems see the sell-off and sell their stocks and similar stocks. This sort of interaction could happen in a second, and no human would ever be aware it even happened.

the way every complex system works is that you deal with problems as they come up. something becomes too onerous to ignore and then you fix it. acceleration & iterative deployment has been the only option: a “pause” in ai development would be entirely squandered