In partnership with
Holiday Creator Calendars Are Filling Up. Q4 Panic Is Optional.
Creators lock in their holiday content calendars 90 days out, before most ecommerce brands finalize their commission strategy and way before Black Friday and October deal events.
Get ahead of the seasonal rush with The 90-Day Holiday Sprint, a practical guide for brands that want creators driving holiday demand while competitors are still recruiting:
Structure commissions by lifetime value, not just first-order margin
Lead with the right products so creators promote with confidence
Recruit and onboard creators with a day-by-day plan for the first 30 days
Read performance early and pull program levers by Day 60
Brief creators with a holiday checklist before calendars fill up
Your 90-day countdown starts now.
Beginners in AI
Good morning and thank you for joining us again!
Welcome to this daily edition of Beginners in AI, where we explore the latest trends, tools, and news in the world of AI and the tech that surrounds it. Like all editions, this is human curated and edited, and published with the intention of making AI news and technology more accessible to everyone.
THE FRONT PAGE
OpenAI Says a Private Model Solved a $1 Million Math Problem, and Nobody Can See It
TLDR: AI has pushed into one of math's six million-dollar problems, and the mathematicians involved now disagree about what was proved and by whom.
The Story:
The Navier-Stokes equations describe how fluids move, from water going down a drain to wind in a storm. For two centuries nobody could prove they always behave, and the Clay Mathematics Institute put up $1 million for an answer. Monday night, NYU mathematician Tristan Buckmaster and Anthropic researcher Levent Alpoge released proofs on closely related fluid equations. They used Claude for the math and checked every step with Lean, software that verifies proofs. In a statement posted with the results, Buckmaster wrote that OpenAI told him one of its internal models had proved the harder case in about 100 pages. OpenAI hasn't named that model or released the proof, and didn't respond to Scientific American's request for comment.
Its Significance:
Two years ago no model could do research math at this level. Now one has helped close proofs that specialists expected to take decades, with software confirming the steps instead of colleagues taking them on trust. Nobody has settled how to credit a model on a paper, or how a journal checks a proof that sits inside a private company. And the prize may go unclaimed anyway, since the method leans on a piece of the equations most mathematicians leave out.
QUICK TAKES
The story: Northwell Health tested an FDA-cleared tool from Aidoc across 3,856 CT scans, published in the Journal of the American College of Radiology. The AI flagged 55 real aneurysms the radiologists hadn't reported, a 39% jump in detection.
Your takeaway: Aneurysms often show no symptoms until they burst. The AI spotted more of them, but doctors were right more often when they called one, so the win comes from running both.
The story: Google and Cathay Pacific ran a trial where pilots on more than 80 flights shifted altitude slightly to skip the cold, humid air where contrails form. Google estimates those flights cut contrail warming by roughly 40%.
Your takeaway: Contrails are the white streaks behind planes, and they cause about a third of aviation's total climate impact. No new planes needed, no new fuel, just better guesses about where the bad air sits.
The story: The French lab closed a 3 billion euro round (about $3.5 billion) at a valuation over 21 billion euros, led by Samsung alongside the EU-backed Scaleup Europe Fund. Mistral says it's the largest equity round a European tech company has ever completed.
Your takeaway: Countries keep picking a home model they can control, and we covered South Korea making the same move a few editions back. Mistral is that idea with a much bigger check behind it. The servers will run on Nvidia chips designed in America.
TOOLS ON OUR RADAR
⏱️ Toggl Track Freemium: Track time across projects with one click, free forever for teams up to 5 people, no trial clock ticking down.
🎙️ Buzz Free and Open Source: Transcribe audio and video entirely offline using OpenAI's Whisper model, nothing leaves your computer, no account, no upload limits.
🎧 Endel Freemium: Get adaptive soundscapes that shift with the time of day to support focus, sleep, or relaxation, free sessions run up to 10 minutes each, start as many as you like.
🗂️ Unclutter Paid: Keep a hidden drawer of clipboard history, quick notes, and loose files one gesture away at the top of your screen, one-time $19.99 purchase, Mac only.
TRENDING
Scientists Draw a Line on Where AI Belongs in Research
In a PNAS Nexus editorial, Charles Branas and Bruce Levine say AI is fine for sorting data and copy editing, and should be disclosed when it helps come up with research questions. What they'd rule out entirely: using AI for the interpretation, the ethics calls, and the policy recommendations. Their word for that is abdication.
ChatGPT May Soon Copy Your Writing Style From Gmail and Slack
OpenAI is testing a Writing Style setting that studies your emails, messages, and docs across Gmail, Slack, Google Drive, and SharePoint, then writes drafts that sound like you. It's for paid ChatGPT Work accounts, and you turn it on under Settings, Personalization, Writing style. Read what you're connecting first, because it reads everything in there.
PCMag Tested the Top AI Chatbots
ChatGPT holds the Editors' Choice spot, Gemini takes best value, and Claude gets credit for a clean interface. Perplexity does web search well but comes up short on deeper research, and Grok has some odd extras. Skim it before you pay for a subscription you don't need.
OpenAI Says Its AI Now Does Research Work for OpenAI
The company says it hit its September goal of an automated research intern, meaning a system that finishes defined tasks that would take a skilled researcher a few days. By mid-August, OpenAI counted 3.1 agent workdays for every human workday across its research org. Next target is a full automated researcher by March 2028.
Sonos Speakers Can Now Take Orders From ChatGPT
Sonos 27 started rolling out today as a free update to S2 systems. A piece called 27mcp lets any outside AI assistant connect to your setup and play songs, shuffle audio between rooms, and change the volume. It's early access, so check the permissions before you hand your house over.
TRY THIS PROMPT (copy and paste into Claude, Grok, ChatGPT, or Gemini)
🔧 Describe the sound your car's making. Get the real commonly-reported causes, ranked, with an honest read on how urgent it is.
Build a single-file HTML app with vanilla HTML/CSS/JS. The Car Trouble Sound Diagnostic — describe a car sound, get real commonly-reported causes via live web search, ranked by likelihood and urgency. No API key needed.
Aesthetic: dark charcoal (#14161a), steel blue (#5a8cb8) primary with a blue glow top-left, caution amber (#e0a840) secondary with a glow bottom-right. Space Grotesk for headings/cause names, Inter for body, JetBrains Mono for labels/tags.
Form: sound description text input, a "when it happens" dropdown (braking/turning/starting/accelerating/idling/at speed/not sure).
Technical: call the Anthropic Messages API WITH tools:[{type:'web_search_20250305', name:'web_search'}] (works automatically), system prompt as an automotive information researcher (explicitly NOT a mechanic performing diagnosis) who finds real commonly-cited causes from reputable automotive/repair sources via search, ranks most-to-least commonly reported, and assigns an honest urgency level per cause using standard categories ("Stop driving, unsafe" / "Get it checked soon" / "Safe to drive short term") — explicitly required to never give a definitive diagnosis and to always frame this as general information requiring an actual mechanic's physical inspection. Return raw JSON: sound_summary, general_note (1-2 sentences on the system this points to), possible_causes (3-5: cause, likelihood, urgency, explanation), safety_note (2-3 sentences reiterating this isn't a diagnosis, plus specific accompanying warning signs — burning smell, dashboard lights, loss of braking — that mean stop driving immediately).
Render: an overview card with the restated sound description and general note. A ranked list of cause cards, left-border and badge color-coded by urgency (red=stop, amber=soon, green=safe), each showing likelihood tag, urgency badge, and explanation. A prominent red "not a diagnosis" safety card at the bottom with the specific warning-sign guidance.What this does: Describe what you're hearing and when it happens, and it searches the live web for real, commonly-cited causes from reputable automotive sources, ranked from most to least commonly reported. Each cause comes with a color-coded urgency level, stop driving, get it checked soon, or safe short term, based on standard automotive safety guidance, so you know what actually needs attention now versus what can wait for a scheduled visit.
What this looks like:
Own Your Brand's AI Voice
The only platform that designs, licenses, and captures a Branded AI Voice from real, consenting actors—not from scraped data. Trusted by BMW, Superbloom, and Cresta.
WHERE WE STAND(based on today’s news)
✅ AI Can Now: Close a 200-year-old math proof that the best human mathematicians couldn't finish.
❌ Still Can't: Explain it. Buckmaster says the English write-up the models produced was barely readable, so he and Alpoge had to rebuild the steps by hand for other researchers.
✅ AI Can Now: Find brain aneurysms on CT scans that trained radiologists missed, 55 of them across 3,856 exams.
❌ Still Can't: Be trusted on its own verdict. When the AI called an aneurysm it was right 78.2% of the time. When a radiologist called one, 92.7%.
FROM THE WEB
RECOMMENDED LISTENING/READING/WATCHING
Flight of the Navigator (1986) - Movie
A 12-year-old boy vanishes in 1978 and reappears eight years later, still exactly the same age, having no memory of where he's been, until a wisecracking alien spacecraft with its own onboard intelligence turns up and starts trying to help him figure out what happened. A Disney-adjacent cult classic (produced outside the studio, later reissued through it) that gave audiences one of the funnier AI ship characters, it was even voiced by Paul Mall and later Pee-wee Herman himself, Paul Reubens.
Thank you for reading. We’re all beginners in something. With that in mind, your questions and feedback are always welcome and I read every single email!
-James
By the way, this is the link if you liked the content and want to share with a friend.
Some * designated product links may be affiliate or referral links. As an Amazon Associate, I earn from qualifying purchases. This helps support the newsletter at no extra cost to you and Amazon makes a tiny hair less.
