Bill Bill

Last breath feeding the AIs with me genious prompts, tweets, git pushes

[Bill Bill @billypages]

@steventey @OpenAI @romainhuet @simonpfish @karoliskosas Blown away by testing Speech-To-Text APIs! 🎙️ OpenAI Whisper vs. #iOS: the accuracy gap is astounding. 🚀 OpenAI requires full audio uploads, but iOS works in real-time. Who else is eager for a streaming version of #OpenAI Whisper? #OpenAIDevDay #VoiceRecognition

[Bill Bill @billypages]

@DeryaTR_ "So how many people have actually died from synthetic viruses compared with natural viruses killing millions every year?"

Is COVID-19 attributed to synthetic or natural?

[Bill Bill @billypages]

@libremax_off on vous prépare à l'identification numérique

[andi (twocents.com) @Nexuist]

One of my deepest insecurities as a man is what if I’ve already been investigated by an intelligence agency and their file on me basically says this dude is lazy and won’t pull off anything important so don’t worry about it

[Bill Bill @billypages]

@BetterCallMedhi Pendant ce temps j'espère que mon dernier prompt Claude alignera la div là où je veux...

[Bill Bill @billypages]

@ChristopheGini Ce n'est certainement pour créer des jeux qu'il a racheté EA Games. Avec un seul prompt à une IA tu peux cloner un jeu. Il doit y avoir un autre deal derrière.

[Bill Bill @billypages]

@thesupermanmx I go to bed by 10:00 PM because I can't find the light switch when it's dark.

[vittorio @IterIntellectus]

while people are still talking about AI, we are already getting closed-loop recursive self-improvement in biology.

a model predicts which mutations could improve an enzyme. an automated laboratory synthesizes, expresses, and assays them. the model learns from the results and repeats the cycle.

researchers at tianjin university ran just five automated cycles and improved one enzyme 57x and another 104x

self-driving biology is here

[Mason @webdevMason]

All that needs to be said about pit bulls was said by S-tier dog trainer Stonnie Dennis in 3 minutes

[Goshawk Trades @GoshawkTrades]

"every single trader out there seems to make the same mistake over and over again. their position size is probably two to ten times more than it should be."

martin shkreli built a kelly criterion simulator to test exactly this.

"most folks don't actually have an edge when they trade. but if they did, they're trading as if they had a 4x or 5x kelly edge."

"if you run the simulator, you will go to zero each time."

"even with a 60/40 edge on every trade you make, you'll go bust if you overbet."

"and who has a 60/40 edge in the stock market? nobody has a 60/40 edge."

even a real edge, sized wrong, still ends at zero.

sizing is too commonly the part I see every trader consider second or third.

[Polymarket @Polymarket]

JUST IN: Scientists propose gradually reducing the global population to 4 billion by 2200, arguing it would “ease pressure" on natural resources & improve quality of life.

[Bill Bill @billypages]

@pandoks Who are the mongols in this parallel?

[Bill Bill @billypages]

The core of this thesis has a giant, obvious hole.

Sequoia (and the people repeating it) are saying:
“Don’t sell a tool that helps the human. Sell the finished work instead. Then every model upgrade helps you instead of killing you.”

But to deliver that finished work you still have to build something — an agent system, workflow orchestration, error handling, integrations, monitoring, etc. That “something” is still software sitting on top of foundation models. When the next Claude/GPT/whatever is dramatically better and cheaper, your differentiation shrinks unless you have real moats beyond “we call the API and slap a services invoice on it.”

The contradiction in plain terms
Sell a pure copilot → model companies eat you.
Sell the outcome via a thin AI wrapper → model companies (or anyone with a better prompt + UI) still eat you.
The only difference is the packaging and the pricing model.

Calling it “the AI accounting firm” instead of “AI for accountants” doesn’t magically create a defensible business. It’s mostly narrative.

Where the thesis isn’t completely empty
There are real advantages if you execute beyond the thin wrapper:

You own the customer relationship and the liability. Customers pay for “books closed correctly” or “contracts reviewed with sign-off,” not tokens. That lets you charge outcome-based prices and keep the upside when models get cheaper.
You accumulate proprietary execution data, edge-case handling, and process knowledge that pure model providers don’t have.
You can start with the messy, already-outsourced work (the low-judgment, high-volume stuff) and gradually push humans further out of the loop.
Trust, compliance, insurance, and audit trails matter a lot in accounting, legal, claims, etc. A random Claude chat doesn’t replace that.

But these are hard operational advantages, not AI magic. They look a lot more like building a scaled services firm that happens to use aggressive automation than like a classic software company. Most founders chasing this “alpha” will just build expensive wrappers that get undercut the moment models improve.

Bottom line
A big chunk of this is classic VC storytelling: reframe the same race against the models as a bold new category so it sounds like insight. The $1 software vs $6 services number is real. The idea that you can cleanly capture the $6 at software margins by just “selling the work” is the baloney part.

The companies that actually win here will look more like ruthless operators who treat AI as a cost-reduction tool inside a real service business — not pure “AI-native” startups that somehow escape the commoditization problem by changing their pitch deck language.

(Grok when prompted)

[Bill Bill @billypages]

@predict_addict She's aware of her struggle compared to her peers.
The question is: was she too smart/creative to bother with competitive math, or was she too dumb for it and diverted to more creative/slow paced math?

[Naval @naval]

It doesn’t matter if open source software is developed by the CCP, the NSA, the Jews, ISIS, MAGA, ANTIFA, or the Illuminati.

It’s open source - just audit and run it yourself.

They can send me gold bricks too, as long as they’re at it.

[Bill Bill @billypages]

@anatolium L’Europe choisit ses alliés avec une remarquable préférence pour ceux qui travaillent contre ses intérêts

[Sammy Obeid @SammyObeid]

Someone send this to Tyler Robinson’s attorney

[Andrew Bridgen @ABridgen]

World Economic Forum founder Klaus Schwab has filed a complaint to police after a listening device was found in his home in Geneva, Switzerland

Ironically, last year Schwab said “We have all got to submit to surveillance, if you have nothing to hide you have nothing to fear”
🤣

[Martin Sonneborn @MartinSonneborn]

Ich habe heute versucht, die Chatkontrolle zu stoppen.

Am Wochenende mussten Sibylle Berg und ich Parlamentspräsidentin Metsola schriftlich mitteilen, dass das Durchprügeln der Chatkontrolle im Eilverfahren leider gegen die Geschäftsordnung des EU-Parlaments verstößt. Während Metsola der interessierten Presse daraufhin (fälschlicherweise) mitteilte, dass das schon alles seine Richtigkeit habe, warten wir immer noch auf ihre Antwort.

Deshalb wollte ich es ihr heute bei der Eröffnungssitzung in Straßburg noch einmal erklären. Und musste erstaunt feststellen, dass die Präsidentin doch Regeln kennt: Exakt nach 60 Sekunden hat sie mir das Mikrophon abgestellt. (regelkonform, wird aber selten gemacht)

Dabei hätte ich noch einiges zu sagen gehabt: "Frau Präsidentin, Sie wachen nach Artikel 22 über dieses Regelwerk (mit der ausgedruckten Geschäftsordnung wedelnd) - erklären Sie den Eilantrag für unzulässig. Wir sind hier schließlich nicht auf Malta! Die aktuelle Fassung der Geschäftsordnung überreiche ich Ihnen gerne persönlich. In der MEP-Bar."

Wie es nun weitergeht? Morgen wird über das Eilverfahren abgestimmt, obwohl diese Abstimmung gar nicht stattfinden dürfte. Wenn der Antrag erfolgreich ist, kommt die Chatkontrolle Donnerstagmittag zur Abstimmung ins Plenum. Um sie noch zu stoppen, müssten 361 Abgeordnete - eine qualifizierte Mehrheit - DAGEGEN stimmen.
Die schlechte Nachricht: Donnerstag ist der letzte Tag vor der Sommerpause und viele MEPs dürften bereits auf dem Weg in den Urlaub sein. Ein Schelm, wer Böses denkt bei dieser Terminierung... Smiley!

Wenn die Chatkontrolle durchgeht, dürfen die Plattformen (also die US-Tech-Bros) wieder & weiterhin fröhlich & ganz legal Ihre Nachrichten scannen. Schreiben Sie also gefälligst etwas unterhaltsamer in den kommenden Wochen... ZwinkerSmiley!

[Bill Bill @billypages]

@brivael C'est deja une star le gars. Il va gagner des fortunes, juste en etant ce qu'il est. Il va faire tous les plateaux tv, twitch etc