
Today's theme is the gap between what AI can do and what anyone promises about it. ChatGPT now builds calculators for you, Claude got a lot cheaper with a line in the fine print, and a McDonald's lawsuit asks who is really setting prices when software makes the suggestions.
Speak naturally. Send without fixing.
Wispr Flow turns your voice into clean, professional text you can send the moment you stop talking. No rough transcription you have to clean up. Actual polished text, ready for email, Slack, or any app.
Speak the way you think. Go on tangents. Change your mind mid-sentence. Flow strips the filler, fixes the grammar, and gives you text that reads like you spent five minutes writing it.
ChatGPT Now Builds You a Calculator. Checking the Math Is Still Your Job.
What happened: OpenAI announced that its new GPT-6 models are rolling out in ChatGPT's Chat tab. Paid users get GPT-6 Sol, and Free and Go users get the lighter GPT-6 Luna. The headline feature is called Intelligent UI, which means ChatGPT decides when a chart, form, button, or calculator beats plain text and builds it right inside the reply. OpenAI says GPT-6 is also better at recognizing the limits of its own abilities, though those performance and safety figures come from its own internal testing.
Why it matters to you: A slider and a clean chart look like real software, and real software feels trustworthy. If you ask ChatGPT for a pricing calculator or a cash flow what-if, the result will look polished whether or not the formula underneath is right. Treat it like a smart intern's spreadsheet: fast and useful, and worth a second look before you quote a customer from it.
What to do about it: Test any calculator ChatGPT builds with a case where you already know the right answer.
Claude Got Up to 90% Cheaper. Here Is the Fine Print.
What happened: Anthropic released Claude Haiku 5.5, its small, fast, budget model. For requests under 100,000 tokens (roughly 75,000 words), it costs $0.10 per million input tokens, which Anthropic says is 90% cheaper than the last Haiku. Past that 100,000 mark the price rises, though Anthropic says it is still 50% cheaper than before. On average, the company says it runs about 75% cheaper. Anthropic also halved the price of cache reads on Sonnet 5.5, its mid-tier model.
Why it matters to you: If you use AI through a tool that charges per use, such as a customer support bot or a document sorter, your vendor's costs just dropped, and some of that should show up in what you pay. Anthropic says about 90% of requests to the old Haiku fell under the 100,000 mark, so the cheap rate covers most everyday jobs. Long, document-heavy jobs are where the bill climbs.
What to do about it: If an AI vendor bills you by usage, ask whether they have moved to the cheaper models yet.
Most marketing events send you home with a notebook full of ideas. This one sends you home with a finished asset. The Epic Marketing Summit is 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. That means you leave with something running, not a framework you still have to go implement. It's built for business owners who want to turn their sales process into a system instead of running it on their own time and energy.
Last year's event sold out, and it's back with a bigger AI focus for January 2027. I wasn't there personally last year and I had major FOMO about it. I know a lot of the trainers, and they weren't talking theory, they'd 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. Seats go faster as the date gets closer. Grab your spot here
McDonald's Is Sued Over an AI Tool That Suggests Menu Prices
What happened: A customer in Illinois filed a proposed class action against McDonald's, as ABC News reported. The suit claims a McDonald's pricing tool shares nonpublic store-level sales data among franchisees who may compete with each other, which it calls algorithmic price-fixing that raised prices. McDonald's says it has used an AI-enhanced pricing tool for over a decade, calls the lawsuit filled with inaccuracies, and says franchisees, not AI, set menu prices. None of the claims have been tested in court.
Why it matters to you: Plenty of small businesses now use software that recommends prices, from hotel rates to contractor quotes. The open legal question is whether a shared tool that learns from many competitors' data crosses a line, even when a human clicks the final button. Nobody knows yet, and it is worth knowing before you adopt one.
What to do about it: Before you buy a pricing tool, ask the vendor what data it pools from other customers.
Suspected AI Agents Hit 7 South Korean Financial Firms and Exposed 68,000 People
What happened: South Korean authorities are investigating a string of bank breaches they believe were likely carried out by AI agents, according to The Record. The personal data of at least 68,000 people was reportedly exposed across at least seven financial institutions, including borrowing history, incomes, phone numbers, and names. Officials believe a Chinese security tool called Artex AI was used, and President Lee Jae Myung said AI has made it possible to hack without specialized skills. The cause is still under investigation.
Why it matters to you: The scary part is not a movie-style super hacker. It is that software can now probe for unlocked doors, like a loan portal that never checks who is calling, in an afternoon and at almost no cost. Small businesses tend to have the same kind of unlocked doors, just with less money behind them.
What to do about it: Turn on two-factor login for your email, bank, and payment accounts this week if you have not already.
Texas Hits Pause on Big New Data Centers
What happened: Texas's grid operator has paused approvals for data centers of 75 megawatts or more to connect to the grid, including 17 projects that had finished every other step, according to an a16z analysis. Requests from large power users in the queue jumped from 63 gigawatts at the end of 2024 to 474 gigawatts by June 2026, about 90% of it data centers. Many of those requests look speculative, and officials worry ratepayers could fund upgrades for projects that never get built. New large-load rules, with a $100,000 study fee and a $50,000-per-megawatt deposit, take effect October 8.
Why it matters to you: AI runs on electricity, and the buildout is now bumping into real limits like power, water, and local patience. For you, that is a reminder that cheap AI is not guaranteed forever, and that power bills are becoming part of the AI conversation in your own town.
The best business advice I've gotten came from a room, not a course. I've been a member of the Ignite Mastermind for over 3 years, and it's been a vital part of how I've grown my business. Some of my closest friends, and a few real business partners, came directly out of being in that room. Ignite is an annual membership for entrepreneurs still under $1 million a year who want structure, tools, and peers to get past that milestone faster.
You get 6 to 8 live virtual workshops a year plus 2 in-person multi-day events, many of which sell for $97 to $997 on their own. There's an all-in-one marketing software suite that replaces over $5,000 a month in separate tools, and weekly calls with Perry Belcher and his team. I also add my 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 what's inside Ignite here
Zuckerberg's Biohub Is Betting $1.8 Billion on a Virtual Human Cell
What happened: Biohub, the science nonprofit backed by Mark Zuckerberg and Priscilla Chan, expanded its Virtual Biology Initiative from a $500 million commitment to about $1.8 billion. The U.S. Department of Energy is putting in more than $500 million over five years, the NIH is sharing existing datasets, and Google DeepMind, Isomorphic Labs, and Meta are adding $300 million together. The goal is open, standardized data to train AI that can predict how cells respond to drugs, often called a virtual cell. The announcement gives no timeline for when the data will be released.
Why it matters to you: This one will not change your business next quarter. It is worth watching because it shows where the biggest AI money is heading after chatbots: science problems with measurable results. If it works, it could shorten how long it takes to test new treatments.
Elon Musk's Grok Bot Will Now Hand Work to Claude
What happened: Elon Musk said Grok Bot, the AI agent app from his company SpaceXAI, will no longer run only on its own models, The Next Web reported. It will route each task to whatever model is likely to give the best outcome, including Anthropic's Claude Opus 5.5, Midjourney, and Suno. Musk did not say which tasks go to which outside model. The announcement came a day after users reported trouble reaching Grok on mobile and web, and the company has not given a cause.
Why it matters to you: The big AI companies are now using each other's models when it helps. That is good news for you: the tool you choose matters less than what you ask it to do, and you are less locked in than the marketing suggests.
The Bottom Line
Look at today's stories side by side. ChatGPT builds you a polished calculator and makes no promise the numbers are right. Claude gets up to 90% cheaper with a threshold in the fine print. McDonald's says humans set prices while a lawsuit says the software does the coordinating. In every case, the technology moved faster than the guarantees around it. That is not a reason to panic or to ignore any of it. It is a reason to keep a human in the loop on anything that touches your money or your customers, check the math, read the fine print, and enjoy the cheaper tools. Stop the rollercoaster, keep your hands on the bar.
Enjoying the Ride?
If this issue saved you a headache, forward it to one person who is trying to make sense of AI. And if someone forwarded this to you, you can 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




