|
|
|
|
Welcome, humans. |
OpenAI just confirmed its most powerful hacking AI is now available through Amazon's cloud store, the same marketplace businesses use to buy everyday software subscriptions. The model, GPT-5.6-Cyber, is trained with fewer safety limits specifically so it can write working exploit code (real attack scripts, not just descriptions of bugs). It's already found over 400 privilege-escalation flaws in a single operating system. |
Until now, getting access to a model like this meant OpenAI personally vetting you. Now it's provisioned the same way you'd add a project management tool: through your company's existing Amazon account, built for speed, not scrutiny. |
Hey! We're booking out ad inventory for Q3 and there's only a few slots remaining! Make sure you reach out ASAP if you want to advertise your product and service to our 700K readers today! |
|
|
P.S: want to get ahead of the crowd? Email Garrett and he'll get you set up right now. |
Here’s what happened in AI today: |
😺 Anthropic's CEO denies wanting to be the only company left. 📰 OpenAI's new ChatGPT feature logs your keystrokes on Mac, unencrypted. 📰 Anthropic is in talks to buy startup Decart for $6 billion. 📰 Stripe is buying AI startup OpenRouter for $7 billion. 🎓 Fine-tune your own AI model using Unsloth Studio.
|
|
|
😻 Anthropic's CEO Got Accused of Wanting to Rule the World. His Reply Was Bigger Than the Accusation. |
Twitter beef between billionaires is usually just noise. This one accidentally became the most honest debate about AI regulation all year. |
It started Friday on the All-In podcast. Investor Gavin Baker said trusted sources told him Anthropic CEO Dario Amodei believes his company could someday be the only private company left standing, with just Anthropic and governments remaining. Baker called it "hubristic," comparing it to SBF-level founder delusion (SBF is Sam Bankman-Fried, the FTX founder now serving time for fraud). |
Here's what happened: |
Anthropic researcher Sholto Douglas called the claim “completely false” on X, arguing Anthropic actually worries most about any one company gaining too much power, not too little competition. Baker didn't back down. He argued Dario's years of warning about AI's risks, like bioweapons or mass job loss, have backfired, fueling today's backlash against AI data centers and killing momentum for federal regulation. Dario Amodei replied directly, in a two-part post that hit 10 million views within a day.
|
|
|
First, he rejected the idea that AI safety only has two options: heavy regulation that hands power to a few giant companies, or no regulation at all. He pointed to California's SB53 (a transparency law Anthropic supported) as proof his policy asks are built to burden frontier labs like Anthropic more than smaller competitors. |
Second, he pushed back on the idea his own warnings caused the public's AI anxiety. “I think it is fundamentally a crisis of trust,” he wrote, arguing people are wary of AI not because of what he's said, but because companies have spent decades overpromising and underdelivering. |
He also shared something personal: his father died of Hepatitis C a few years before sofosbuvir, a drug that now cures 95% of patients, became available. It's part of why curing disease with AI matters so much to him. |
Why this matters: You've probably felt this tension yourself. You use an AI tool like Claude or ChatGPT most days at work, and it's useful. Some part of you still doesn't fully trust it with your data, your job security, or where any of this is headed. Call it the "crisis of trust" Dario's describing, just showing up in your own inbox instead of a Senate hearing. The real question is whether any AI company, Anthropic included, can earn back your benefit of the doubt. If not, that distrust hardens into rules that change what tools you're even allowed to use at work. |
|
|
FROM OUR PARTNERS |
|
|
With AI, non-technical employees are shipping apps, agents, and automations faster than IT can track. The best companies are letting everyone build, without putting their core operations at risk. |
On August 19, join Tines to hear from Headspace's Director of AI Enablement for an exclusive look at how Headspace enables everyone to build securely with AI at scale: |
How Headspace governs wild code at scale: gaining visibility into tools built outside IT's radar. Giving teams the freedom to build without operational risk: adding governance and control, without slowing innovation. Monitoring AI-built workflows in practice: understanding what’s running, how it’s performing, and where it's spending.
|
Save your spot |
|
|
🎓 AI Skill of the Day: Fine-tune your own AI model |
Sometimes a general-purpose AI knows plenty, but doesn’t behave the way you need. Maybe you want a model that writes like your company, understands a specialized workflow, or tutors you at exactly your skill level. |
That’s what fine-tuning does: you take an existing open-source model (meaning a model you can download, run, and customize yourself) and train it on examples of the answers you want, essentially teaching it new habits without building an AI from scratch. |
Unsloth Studio makes the technical part much easier. Think of it as a point-and-click workshop for customizing AI models on your own computer: choose a model, feed it examples, train it, test the results, and export your custom version. |
The basic workflow: |
|
|
Then comes the important part: compare your fine-tuned model against the original to see whether your data actually made it better. |
The secret sauce is giving a smaller model really good examples of exactly what you want it to do. Basically: start with a smart generalist, then teach it to become your specialist. |
Have a specific skill you want to learn? Request it here. |
|
|
🍪 Treats to Try |
*Asterisk = from our partners (only the first one!). Advertise to 700K+ readers here! |
*AI is here, but data architecture is still catching up. See why 72% of enterprises are planning an overhaul. Download the report now Zetik watches anything you name (a company, a person, a rumor) across the entire internet, and only pings you when something actually happens. Vocal Slice lets you cut a podcast or interview clip just by highlighting the words you want in a transcript, no audio editor required —free trial, then $29/year. CostLogic turns a construction blueprint PDF into a measured takeoff, a priced estimate, and a sent invoice, all in one browser tab. Blume turns a folder of plain markdown files into a full, searchable documentation site with a single command —free to try. Mole researches a question for you on a hard dollar budget you set upfront, and throws out any claim it can't verify word-for-word against its source —free to try. claudeconfirm makes you type a confirmation before it'll let you open Claude, so you catch yourself reaching for AI out of habit —free to try.
|
|
|
📰 Around the Horn |
 |
The watermark got explained. The eraser got built. An open-source tool that strips AI watermarks out of Claude, Gemini, and OpenAI content just hit 11,000 GitHub stars, days after Anthropic detailed how Claude's own watermark works. |
|
Stripe finalized a deal to acquire AI model marketplace OpenRouter for more than $7 billion, over five times the $1.3 billion valuation OpenRouter had just three months ago. An unreleased Anthropic model made real progress on the Riemann hypothesis, a 150-year-old math problem with a $1 million prize still unclaimed. Hugging Face's new report found Chinese labs are now releasing open AI models several times larger than anything U.S. labs have shipped this year. OpenAI launched a ChatGPT feature that logs every click, keystroke, and app switch on your Mac and stores it as unencrypted plain text. Anthropic is in talks to buy AI infrastructure startup Decart for about $6 billion, which would be its largest acquisition ever.
|
|
|
FROM OUR PARTNERS |
200 Ways To Make Money With AI |
|
|
Ready to transform artificial intelligence from a buzzword into your personal revenue generator? |
HubSpot’s groundbreaking guide "200+ AI-Powered Income Ideas" is your gateway to financial innovation in the digital age. |
Inside you'll discover: |
A curated collection of 200+ profitable opportunities spanning content creation, e-commerce, gaming, and emerging digital markets, each vetted for real-world potential Step-by-step implementation guides designed for beginners, making AI accessible regardless of your technical background Cutting-edge strategies aligned with current market trends, ensuring your ventures stay ahead of the curve
|
Download your guide today and unlock a future where artificial intelligence powers your success. Your next income stream is waiting. |
Get The Free Guide |
|
|
|
😹 Monday Meme |
 |
Too funny not to share. Codex-pets.net lets you pick different Codex pets, including one modeled on Anthropic CEO Dario Amodei! There are plenty of other cute picks too. Special shoutout to Microsoft's Clippy. |
|
|
|
New from The Neuron: AI Explained |
|
|
New episodes air every week on Wednesdays: Spotify | Apple Podcasts | YouTube |
P.S: We’re trying to hit 50K subscribers on YouTube this year. Click here to help! |
|
|
|
|
|
|
|
|
That’s all for now. If you want to get featured above, fill out the poll below and tell us how we did today! |
|
What'd you think of today's email? |
|
|
|
|
Btw: We just launched a robotics newsletter! Sign up for it here. |
P.S: Love the newsletter, but only want to get it once per week? Don’t unsubscribe—update your preferences here. |
No comments:
Post a Comment
Keep a civil tongue.