Alpha_🜲

Anthropic and OpenAI are going after two different giants, they just dropped big launches 40 minutes apart today. Claude went after Microsoft Office with Sheets, Whiteboards and Data ChatGPT went after Google it's called Businesses in ChatGPT. you know how restaurants and shops claim their page on Google so they control what shows up when you search them? same thing but inside ChatGPT businesses can claim their page, fix their info, add products with current prices and see how often ChatGPT mentions them and some US online stores can now build their own Business Agent, basically a chatbot for their shop that answers your questions right inside ChatGPT if your business isn't on ChatGPT soon it's gonna be like not being on Google Maps in 2015 OpenAI says these business chats will be clearly labeled and kept separate from ChatGPT's own answers let's see how long that lasts because OpenAi once tried to go after Google with a browser that miserably failed, do you think this one is it?

They actually built an AI Phone no not an iPhone an actual “AI” Phone a YC startup called Blue says they switched 10 people from iPhone to their AI phone and none of them want their iPhone back 10 people. that's not a study bro that's a group chat 😭 but the idea is actually interesting Blue is a phone you mostly just talk to. you tell it what you need and its AI goes into your apps and does it, replying to messages, comparing prices, doing expense reports, even handling your 2FA codes the phone right now is a Google Pixel 10 Pro running their own version of Android. their own hardware isn't coming till 2027 it's $1,299, or basically free if you trade in your iPhone within 30 days (they charge you first and refund you when they get it). after 3 months the AI is planned to cost from $20 a month one thing tho, the video says it triages your notifications but on their own site that's listed as "coming soon" for the beta, same with the "Hey Blue" wake word so right now it's kinda a Pixel in a trench coat with a really smart assistant inside and yes, US only ✌️🤣😂🤣😂🤣😂 I swear everything starts from the US

Polymarket says Tim Cook is "reportedly" being floated to replace Jensen Huang at Nvidia but let's be real here this is only Polymarket. not a single major outlet is reporting the Tim Cook part at all... and Polymarket literally runs a betting market on who Nvidia's next CEO will be so... big spoon of salt on that one but that doesn't mean Jensen isn't leaving. they're actually planning to kick bro out. according to the Wall Street Journal, Nvidia's board is preparing to push Jensen out. the New York Times reported a second relationship, this time with an Nvidia VP who reports directly to him, and Bloomberg says the board is working with an outside firm to find a replacement and might name a temporary CEO first the stock dropped almost 8% after hours... life ain't easy in the corporate world man 3 days ago the board said it still had confidence in him. that confidence lasted about as long as a Snapchat story fr 😭 the man started Nvidia in a Denny's booth in 1993 and turned it into one of the most valuable companies on earth. imagine losing it like this. this is why CEOs like Elon set things up so nobody can ever fire them man😂🤣😂🤣😂😂🤣 THE GOAT ELON cause how can they do this to him?

6 days after Claude launched Personal Finance, ChatGPT just dropped Finances on desktop it helps you find ways to save, see where your money goes, make sense of your investments and even build your own little money tools and here's the difference Claude's version is only on paid plans, it's read-only and Anthropic says it's not built to give investment advice ChatGPT's goes to Free users too and goes straight into your investments so Anthropic went careful and OpenAI went for everyone. same idea, same week, different energy these two are basically copying each other's homework at this point and of course it's US only again, what a surprise right?

“Gemini 4 Carbon, now available” yeahhhh “to Ultra subscribers”

Anthropic just gave paid Claude users a free extra life 😭 one full reset of ALL your usage limits across every model, for Pro, Max, Ultra and Team. you pick when to use it, anytime before Dec 10 why? they say they fixed some small issues that were eating more of your limits than they should pro tip: don't use it right now if you're not close to your limit. save it for the day you're deep in a project and hit the wall you'll find it in Settings → Usage a few days ago the top reply on their subagent update said "kind of sounds like you fixed a bug". well... 👀 Google is locking Carbon behind $100 a month and Anthropic is handing out free resets. pick your fighter 😭😭🫶

Google finally made Gemini 4 Carbon available, 2 weeks after they "dropped" it and only let cybersecurity teams touch it and Google came in swinging on price. through Nov 30 it's $2 per million input tokens and $10 output. that's a quarter of what Claude Fable 5.2 costs and half of Opus 5.5 then on Dec 1 the price doubles to $4 and $20. Google basically ran a Black Friday sale on AI 😭 is it better than Fable tho? not really. Fable still wins most of the big tests. but Carbon beat it on a couple, like Terminal Bench 4.0 and an automation test where it got the job done for about half the cost per task they also previewed /swarm in Antigravity. one prompt turns into a whole team of agents and Carbon plays the boss but if you just want it in the Gemini app you need Ultra, starting at $99.99 a month. Pro users in the replies are already threatening to cancel and I don't get it. Google hasn't had a model that blew everyone's mind in a while. the last time they really cooked was a Pro model a while back. so why lock a model that isn't even beating the competition behind a $100 paywall? if you want people to pay, let them get their hands on it first and see how they react. Anthropic put Fable 5.2 on every paid plan on day one and the whole timeline was talking about it it's starting to feel like Apple dropping a new iPhone with minimal improvements, get serious Google

Claude is quietly turning into Microsoft Office and nobody's saying it out loud Just look at what they rolled out today 1. Whiteboards, you talk through a messy plan and Claude lays it out on a board your team can edit 2. Sheets, ask for a budget or a tracker and you get a real working spreadsheet everyone can keep building on 3. Data (used to be Dashboards), turns your numbers into dashboards, data stories and even narrated videos and you can now open Adobe, Atlassian and Box right inside Claude and work in them side by side Excel has been around since 1985 and I still don't know how to make a pivot table. I might never have to now, truly all those years of refusing to learn Microsoft products are paying off hallelujah 😭😭😭😭😭🫶🫶🫶🫶🫶 but heads up, Whiteboards and Sheets are still in beta and only on paid plans so free users are watching this one from the outside

Cheating as a huge public figure is the worst thing you can do man😭😭😭. The things I have seen on this app because of the affair Jensen had with the Asian baddie are out of this world

Anthropic just dropped a guide on which Claude model to use since they dropped Fable 5.2 here is what you need to know. think of it like a hospital. you don't call the surgeon to put on a band aid - so pick Haiku 5.5 the cheap fast one for simple repeat stuff like sorting, tagging, pulling info out. on short prompts it costs about 1/20 of Sonnet - then go for Sonnet 5.5 for everyday stuff with clear instructions, bug fixes, chat, docs, slides - and our beloved Opus 5.5 the default. real engineering, debugging, anything you're not sure about - and finally the beast Fable 5.2: the surgeon. huge refactors, planning big systems, deep research. costs twice as much as Opus The biggest tip is before you jump to a bigger model, turn up the effort first. and if a model keeps failing, make sure it can check its own work with a test before you throw more money at it also put your subagents on Haiku for the boring stuff like reading logs and searching files. Anthropic literally says that so yeah a few days ago I asked who actually uses Claude Haiku. turns out Anthropic says your subagents should, my bad for hating on Miss Haiku yeah it’s a she and the funny part, half the timeline switched everything to Fable yesterday but Anthropic's own advice is to start with Opus 5.5 on medium ✌️

Sometimes I wonder will Gemini ever get it’s moment ever again…. It seems like every model Google releases gets obliterated before it can even be publicly available

This will clap out the usage limit easily but seems like a good idea

Claude Fable 5.2 built a working home voice assistant from ONE paragraph. not the code, the whole thing it wrote the specs, picked the parts, designed the circuit boards, designed a case you can 3D print and wrote the software. boards were ready to manufacture in under a day and all three devices it built worked without fixing anything I mean it’s anthropic at this point they can probably do anything it also wrote a full plan for making 4 complex molecules from just a picture of each in about 30 minutes, and experts said it's early grad student level work and it solved a black hole physics calculation that had been sitting unsolved for 23 years but before everyone loses their mind, the chemistry hasn't been tested in a real lab yet and the physics papers haven't been reviewed by other scientists yet so calm down... might be just hype for the release man

Anthropic just dropped Claude Fable 5.2 and Mythos 5.2 and half the timeline is about to tell you Claude solved Fermat's Last Theorem. it didn't. let me explain Fermat's Last Theorem is a math problem that took humans over 350 years to prove. what Anthropic did was put 1,000 Fable 5.2 agents on it and in 30 hours they translated the whole existing proof into a language a computer can check step by step Opus 5.5 tried the same thing, got 26% done and then basically stopped moving 😭 so which one do you actually use? Fable and Mythos are the same model. Fable is the one with strong safety guardrails and you can use it today on any paid plan. for everyday coding Anthropic still says start with Opus 5.5 Fable also got cheaper, $8 per million input tokens and $40 output, down from $10 and $50. and at high effort it codes about as well as Opus 5.5 at max effort for roughly half the cost and Mythos? only cybersecurity and medical research orgs that get approved can use it so forget about it ✌️

They made a video 😂😂😂😂😂😂

Jensen Huang, the man whose chips literally power all of AI, just admitted to cheating on his wife and bro this is not regular celebrity gossip he sent a letter to Nvidia staff saying "I was unfaithful to my wife" and "there is no excuse for what I did" but here's where it gets crazy, US intelligence reportedly believes the woman worked for the Chinese government. she was an exec at a company that buys Nvidia chips and is accused of sneaking about $2B worth of them into China China says it's all lies, Nvidia says no secrets were shared and the board is now investigating with outside lawyers the US spent years building walls to keep Nvidia's best chips out of China and if this spy thing turns out true the way around it wasn't hacking, it was the CEO 😭 anyway he's staying CEO and Nvidia says this "will not slow Nvidia" down GPU prices still not coming down tho ✌️

Plus users

that's like a kid who learned to cook from his mom then saying moms were never good at cooking 😭 I get the point but I hate how this makes it sound like the OGs were irrelevant humans wrote the code that led to the creation of the models that are now writing better infrastructure. you can't take nothing from the creators those 588 new Linux bugs GPT-6 Cyber found yesterday? they're in a kernel humans started building in 1991 that now runs most of the internet and every Android phone. 30+ years of "not very good" keeping the whole world online and the models only know how to find those bugs because they learned from code and security research humans wrote humans weren't bad at this. they built the thing that's now better at it and when the "turbulence" hits, guess who's still gonna be on call at 3am fixing it ✌️