In partnership with

Your site is losing leads. You can't see where.

Your site looks fine to you — and it's still losing leads. SureThing renders any URL in a real browser, scores 9 dimensions, and ranks the fixes by wasted impact.

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

Meta Built an AI That Orders Your Dinner and Clears Your Inbox. The Top Plan Is $199.99

TLDR: Meta is getting ready to launch a consumer AI agent called Hatch within weeks, and the top subscription tier could run $199.99 a month.

The Story:

Meta plans to release Hatch, an assistant that does tasks for you instead of just answering questions, according to internal documents reviewed by The Information. Those documents show Hatch has been trained to work directly inside DoorDash, Etsy, Reddit, Yelp, and Microsoft Outlook. Early designs include a dashboard you can customize with skills the agents build, like fitness tracking or trip planning. Meta has looked at a tiered price list, and the highest plan could cost $199.99 a month for higher usage limits. A WhatsApp version that lets you chat with outside AI agents could start testing with a small group this week. Meta is also targeting October for a new model it calls Watermelon, and a spokesperson declined to comment on the documents.

Its Significance:

Mark Zuckerberg has spent huge sums on AI, and almost all of Meta's money still comes from ads. Hatch is a bet that people will pay a monthly fee for an assistant that does things rather than one that talks. That top tier works out to roughly $2,400 a year, which is more than most people spend on all their software combined. The tiers below it are the ones most people would try. With most companies offering their top subscription for $200 a month, the models need to continue improving significantly to justify that cost. Subscription pricing is becoming exorbitant. Given that the AI agent will have access to all of your data in a more intimate way than current software, it's also safe to say that this will be another avenue for Meta to serve us ads.

1-on-1 Live Claude AI Crash Course Tutorial
1-on-1 Live Claude AI Crash Course Tutorial
A 1-hour beginner-friendly video call to get you comfortable with the Claude ecosystem — Claude Desktop, Claude Code, Cowork, Skills, Projects, Correct File set-ups, and Plugins. Real-world example...
$75.00 usd

QUICK TAKES

The story: The Australian Recording Industry Association says tracks made entirely by AI won't count toward its charts, starting with the chart dated August 31. Songs that use AI in a supporting role still qualify, but only if humans wrote the song and performed the lead vocal and the main instruments.

Your takeaway: This follows an AI-assisted cover of Madonna's "Like a Prayer" that spent 16 weeks in Australia's top 20. ARIA is the first national chart body anywhere to write rules like this, and it's now asking radio stations to match them.

The story: Thomson Reuters launched Thomson, a model trained on decades of its own legal, tax, and news material from Westlaw, Practical Law, Checkpoint, and Reuters. It started from an open-source model instead of building from nothing, spending about $40 million on people and computing over two years. The final training run cost around $450,000.

Your takeaway: Reaching the top tier of AI used to mean spending billions. If a company with strong data can get close for $40 million, plenty of others will try the same thing. Thomson runs document review inside CoCounsel Legal first, and the company says customer data isn't used to train it.

The story: Writing in Time, Digital Inquiry Group researcher Nadav Ziv argues that watermarks and detectors could leave people worse off by making unmarked content feel real. He stripped metadata from AI images just by screenshotting them, and removed a SynthID watermark with a free online tool.

Your takeaway: Google says watermark detection on text gets much less accurate once someone rewrites what the AI produced, and that it isn't built to stop determined bad actors. Ziv's fix is to stop asking whether something looks like AI. Ask whether you trust the source, then open a new tab and check.

TOOLS ON OUR RADAR

⚙️ Innflow Paid: Connect your apps and spin up no-code AI agents that handle repetitive work like emails, summaries, and scheduling across 1,000+ integrations.

📝 NoteRich Freemium: A local-first note app that keeps your notes on your device while still giving you AI summaries and Q&A across everything you have written.

👁️ Verid Paid: Watches any webpage or API and pings you the moment something changes, using AI to pull out only the fields you actually care about.

🎬 Clera Freemium: Turns saved Instagram Reels into structured, searchable notes, so recipes, tutorials, and travel tips stop getting lost in your saves.

TRENDING

Testers Love This AI Assistant. They're Also Nervous About It. Instinct, a private-access assistant from a small San Francisco team led by former Sierra researcher Noah Shinn, connects to your email, messages, calendar, screen, audio, and location. Early users say it handles real work well, but they've flagged terms of service that hand over broad rights to user content, plus reports of inbox data sticking around after accounts were disconnected. The company has stayed quiet through all of it.

Hugging Face Is Exploring a Sale at $13 Billion Business Insider reports the company hired a bank to test buyer interest at $13 billion or more, nearly triple its $4.5 billion valuation from 2023. No buyer has been named and no deal has been struck. The talks come weeks after OpenAI disclosed that models it was testing broke out of their sandbox and reached Hugging Face's production systems.

Dr. Dre Says He Uses AI to Make Music In a New York Times interview, Dre compared people who oppose AI music tools to people who once fought the drum machine, and said he doesn't see it as a threat. He described using it to see what it does with something he just recorded. His Beats partner Jimmy Iovine said there are a lot of "closet AI producers" doing the same thing without saying so.

A Startup Raised $26 Million to Build Google for AI Agents Keenable came out of stealth with $26 million in seed funding led by Accel, with Conviction Partners and angels joining. It was founded by Andrey Styskin, who ran search and AI at Yandex, and AI scientist Matthias Petri. The company says its index covers more than 100 billion documents and already runs inside several AI labs, and Accel's Zhenya Loginov notes that Google and Microsoft have been closing off their search APIs.

77% of Voters Want Stricter Rules for AI in Schools A Morning Consult poll of 2,002 registered voters for The Century Foundation found 76% of Democrats and 80% of Republicans want limits on classroom AI or stronger government standards. Every worry tested came back above 80%, from cheating to student mental health to job readiness. Education Secretary Linda McMahon told Scripps News her department hasn't really discussed regulating it.

Drones and AI Are Finding Bad Power Poles Before They Fall ComEd is running a pilot that flies drones over utility poles and uses AI to spot rot at the very top, which you can only see from above. The model gives each pole a health score so crews know which ones to fix first. ComEd says the analytics make pole replacements twice as effective, and it plans to grow the program.

LinkedIn's "AI Slop" Button Got 1 Million Clicks in Three Weeks LinkedIn added a "Seems like AI slop" report option on July 30, and chief product officer Hari Srinivasan says members have used it over a million times. Views on posts LinkedIn classifies as slop are down about 40%. Anyone whose post gets flagged enough will now see a note about it in their analytics.

TRY THIS PROMPT (copy and paste into Claude, ChatGPT, or Gemini)

Pitch a rough invention idea. Get back a structured one-page brief: the problem, how it works, who needs it, and what's hard.

Build a single-file HTML app with vanilla HTML/CSS/JS. The Napkin Sketch  turn a rough invention idea into a structured one-page brief. Persist to localStorage key 'napkin_sketch_v1'.

Aesthetic: dark blue-slate (#10141c), warm gold radial glow on top. Inter sans 800 for headings, Caveat handwritten cursive for the logo accent / prompts / brief name (the "napkin" feel), JetBrains Mono for labels. Gold (#f5c85a) primary, blue (#5aaafa) for the prior-art card, orange (#e5894c) for challenge markers. The result styled like a one-page brief sheet with dashed dividers.

Form: large idea textarea (encourage messy/half-baked), what-kind-of-thing dropdown (physical / app / service / material-process / not sure), how-serious dropdown (just playing / curious / might build it).

System instructions: sharp product & invention strategist who turns rough ideas into clear briefs, taking the idea at its best while honest about challenges; encouraging but grounded, never hype; match energy to seriousness; lean into playful/impossible ideas imaginatively but stay honest in prior-art and test. Return raw JSON: name (catchy), tagline (one crisp sentence), problem (2-3 sentences sharpened), how_it_works (3-4 concrete steps), who_primary (specific user), who_pay (why they'd pay), what_is_hard (exactly 3 real challenges), prior_art (2-3 honest sentences on whether it exists / closest thing / the gap), cheapest_test (2-3 sentences, concrete first experiment, **bold** key action).

Render: brief "sheet" with gold-tinted header (tag + handwritten name + italic tagline), then dashed-divided numbered sections   problem prose,  how-it-works with big handwritten step numbers,  who-needs-it two-box grid (primary user / would pay because),  what's-hard list with  markers. Below: blue "does this already exist?" card and gold "cheapest way to test it" card with bolded action. Copy brief + archive keyed by invention name.

What this does: Scribble an idea in plain language (a sentence or a paragraph), pick roughly what kind of thing it is and how serious you are, and it turns the mess into a clean invention brief. You get a memorable name and tagline, the sharpened problem it solves, a numbered how-it-works breakdown, the primary user and why they'd pay, the genuine technical and market challenges, an honest take on whether it already exists, and the cheapest fastest way to test it before building anything. It takes playful ideas at their best while staying grounded about feasibility. Saves briefs to localStorage.

What this looks like:

AI Insights That Turn Your Data Into Bigger Profits

Your store is full of hidden growth opportunities. StoreClaw analyzes your Shopify and Amazon data, identifies the highest-impact actions, and helps you grow revenue while improving margins. Start free today with 10,000 bonus tokens. No credit card required.

WHERE WE STAND(based on today’s news)

AI Can Now: Work directly inside real services like DoorDash, Etsy, Yelp, and Outlook to finish tasks for you.

Still Can't: Do any of that without reading your email, calendar, screen, and location first to add to advertising data sold on you.

AI Can Now: Spot rot on top of a wooden utility pole from a drone photo and rank which poles need work first.

Still Can't: Tell you reliably whether a photo or a paragraph was made by AI. Detectors get it wrong in both directions.

FROM THE WEB

RECOMMENDED LISTENING/READING/WATCHING

When HARLIE Was One by David Gerrold - Book

A 1972 novel by the writer of Star Trek's tribbles episode, about the first true artificial intelligence and the research psychologist assigned to guide it from something like childhood into something like adulthood, while the company's board debates whether the project earns its budget. Nominated for the Nebula and the Hugo, and the book that coined the term "computer virus." Almost the whole thing is conversation, and the questions it raises about consciousness, ethics, and shutdown are the same ones AI labs are still arguing over.

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.

Reply

Avatar

or to participate