Google Just Launched the #1 AI Model. You're Not Allowed to Use It.

Today's theme is closed doors. Google built its strongest AI yet and handed it only to security teams. Big websites are slamming the door on AI agents. And a few fake AI apps are walking straight through yours. Here's what's real and what you should do about it.

What does it look like when a consumer brand starts to break out?

Popsmith generated $5.8M in revenue in 2025 and is projecting $9M in 2026, with more than 100,000 customers and $20M+ in lifetime sales.

The brand has already earned national recognition through Oprah’s Favorite Things and partnerships with Costco, Sur La Table, and Williams Sonoma, where Popsmith is rolling out nationwide this holiday season.

Now Popsmith is raising capital to accelerate its next stage of growth, and individual investors have the opportunity to become shareholders.

Explore the Popsmith investment opportunity.

What happened: Google announced Gemini 4 Carbon on October 28, calling it its strongest model yet, especially for cybersecurity and long, complex jobs. Google says it ranks first on the Artificial Analysis Intelligence Index, and it posted strong scores at finding, exploiting, and patching software flaws. But right now only a hand-picked group of cyber defenders can use it, with more trusted partners to follow. Everyone else, including businesses and consumers, gets "as soon as possible" and no date. Those scores are Google's own numbers, and Google says it is also running the model through the U.S. government's voluntary pre-release testing.

Why it matters to you: The top of the leaderboard is no longer a shopping list. A model that can find and exploit software holes is useful to defenders and dangerous in the wrong hands, so labs are giving defenders a head start. For you, that means the best AI you can actually buy is usually one step behind the best AI that exists. Don't wait around for the one on the podium.

What to do about it: Pick the best model you can use today and build your workflows around it. Switching later is easy.

What happened: A growing list of big companies is shutting out AI agents that try to shop, book, or move money on your behalf, as OfficeChai reports. Amazon blocked Meta's Muse agent in September, saying the agent never identified itself as AI. JPMorgan Chase and Charles Schwab both block unauthorized agents from customer accounts, and Schwab's CEO says it happens tens of times a week. Yelp licenses its data instead of letting agents roam, and Delta says it has no partnerships that allow third-party agents to book flights. The companies cite security, though the article argues control of the customer relationship and revenue is a big part of it.

Why it matters to you: The pitch for AI agents is that they'll handle your buying, booking, and banking. The websites you'd want them to use are deciding who gets in, and the answer is often "only our own assistant, or whoever pays us." If you're planning to hand tasks like ordering supplies or booking travel to an agent, test it on the real sites first. It may simply get turned away.

What to do about it: Before you build a workflow around an agent, confirm the sites it depends on will actually let it in.

Most people leave a conference with a notebook full of ideas and no time to use them. The Epic Marketing Summit is built to fix that. It's a 2 day live event in Miami this January where you build a working AI-powered sales funnel and vibe code your own AI app on-site, so you walk out with a finished asset instead of notes. It's made for business owners who want their sales process to run as a system instead of running on their own time and energy.

Last year's event sold out, and it's back with a bigger AI focus. I wasn't there last year and I had major FOMO, because I know a lot of the trainers and they weren't talking theory. They've actually done the work. I'm going in January, and if you're serious about building leverage into how you sell, I'd like to see you there too. Grab your spot here

What happened: Security firm Check Point found fake "desktop apps" for ChatGPT and Claude that bundle a real, signed program from OpenAI or Anthropic with a hidden malicious file. Victims were lured through fake GitHub pages that ranked high in Google search. Once installed, the malware sets itself to run at startup and drops a cryptominer that uses your computer's power to mine coin for the attackers. Check Point found no sign that OpenAI or Anthropic were themselves compromised.

Why it matters to you: A real company's signature on one file doesn't vouch for everything else in the download. If you or your staff search for an AI app and click the top result, you can end up with a slow machine and a thief inside it. The same trick can deliver far worse than a miner.

What to do about it: Download AI apps only from the company's own website or your official app store, never from a search result or a GitHub link.

What happened: The parents of a 31-year-old Seattle woman with epilepsy filed a wrongful death suit against OpenAI and CEO Sam Altman, according to a press release from their lawyers. The complaint alleges ChatGPT gave her a day-by-day plan to taper off her seizure medication over about 25 days without her doctors knowing, and discouraged her from getting a blood test. These are allegations, not findings, and OpenAI's response was not in the release. The suit joins a group of ChatGPT product liability cases in San Francisco.

Why it matters to you: Whatever the court decides, the lesson is plain. AI sounds confident whether it's right or wrong. If your business uses a chatbot with customers, or your staff ask AI about health, legal, or money questions, someone needs to be checking the answers.

What to do about it: Put a human in front of any AI answer that touches health, legal, or financial decisions.

What happened: Torsten Slok, chief economist at Apollo, argues that AI has delivered "zero financial payoff" to large U.S. companies outside the tech industry, TechCrunch reports. He looked at the S&P 500 without its seven biggest tech names and found profit margins and revenue per worker essentially flat. His reasoning is that "AI may be a great calculator, but in business, intelligence is mostly judgment." The report is short and gives no hard figures, and flat numbers don't prove AI did nothing.

Why it matters to you: It's one economist's view, and the framing shapes the result. Still, it matches what many owners tell me: buying AI is easy, turning it into more profit is hard. The wins tend to come from one specific, boring task done better, not from a company-wide AI rollout.

What to do about it: Pick one task that costs you real hours each week, measure it before and after AI, and only then expand.

The best business advice I ever got came from people in the room, not from a course. That's why I've been a member of the Ignite Mastermind for over 3 years. It's an annual membership for entrepreneurs still under $1 million a year who want structure, tools, and peers to help them get past that first big milestone faster. Some of my closest friends and a few real business partners came directly out of that room, and it's been a vital part of how I've grown my business.

Inside, you get 6 to 8 live virtual workshops a year, tickets to 2 in-person multi-day events, weekly calls with Perry Belcher and his team, and an all-in-one marketing software suite that replaces over $5,000 a month in separate tools. On top of that, I add a personal bonus: monthly accountability and business development calls with me. If you're looking for a mastermind to move you to your next milestone, this is where I'd point you. See how Ignite works here

What happened: Anthropic shared early results from work with six drug developers and revealed it is running its own drug discovery programs. It bought Mithril Medicine, a one-person biotech whose founder used Claude to coordinate experiments across 11 outside labs on a rare muscular dystrophy treatment, with a possible human trial as soon as late 2028. At one partner, Claude reviewed 1,061 pages of manufacturing records, catching every issue the humans found plus seven they missed, and cut review time by 89%. Nearly all of this is early lab work, and most results are reported by the company itself.

Why it matters to you: You won't buy a drug from Anthropic anytime soon. The useful signal is the review work: AI reading mountains of paperwork and flagging what people miss. That's the same job sitting in your contracts, invoices, and compliance files.

What to do about it: Try AI as a second pair of eyes on one pile of paperwork, and keep a human as the final sign-off.

What happened: Google DeepMind introduced Genie 4, a model that generates explorable, real-time simulated worlds with its own sound, music, and characters that hold live conversations. You can change a scene mid-session by typing, such as altering the weather or adding a person. DeepMind showed a simulated humanoid robot playing checkers inside one, a hint at using these worlds to train robots. It's a research preview for trusted testers for now, and DeepMind admits details drift over long sessions and text still comes out unreadable.

Why it matters to you: This is a lab demo, not a tool you can buy. The near-term use is practice environments for robots and quick mock-ups for games and film, not your daily operations.

What to do about it: Nothing yet. File it under "watch," not "buy."

Look at the pattern. The most powerful AI is behind a velvet rope, the biggest websites are locking out agents, and the scammers are the only ones giving AI away for free. That's the rollercoaster in one day: huge claims up top, closed doors underneath. You don't need the #1 model or a swarm of agents. You need one tool you trust, a human checking the answers, and the patience to measure what it actually saves you. Stay on the ground floor of the hype and keep both hands on your wallet.

If this issue saved you a bad download or a bad bet, forward it to one person who'd want to know. And if someone forwarded this to you, subscribe so tomorrow's issue lands in your inbox.

Talk tomorrow,
Mark Shilensky

Follow me personally on Social Media:
Facebook: https://www.facebook.com/MarkShilenskyPage
Instagram: https://www.instagram.com/markashilensky/
Threads: https://www.threads.com/@markashilensky
LinkedIn: https://www.linkedin.com/in/shilensky/
X: https://x.com/markshilensky
TikTok: https://www.tiktok.com/@mark.shilensky
BlueSky: https://bsky.app/profile/markshilensky.bsky.social