Sponsored by
Every major AI lab spent the weekend agreeing the industry needs to slow down. By Monday morning, the President had called it a sick conspiracy and told Anthropic's CEO to stop pretending to be an angel. That's the kind of week it's been. Here are the 8 stories that actually matter for you, from a new chatbot law with real teeth to a robotaxi trying to out-cute Waymo.
Smarter CRM. Less Busywork.
Disconnected data and tools make it harder to understand your customers. HubSpot's Agentic Customer Platform brings your data, teams, and tech stack together with AI built in to help your business work faster and create more personalized customer experiences.
Why HubSpot and what's new
Use AI powered tools to take action faster
Unify your data, teams, and tech stack in one place
Create one shared view of customer data
Connect teams around the same customer context
Bring your business tools into one place
Connect more of your business in one place and give every team a smarter way to work. Get set up quickly and start checking off your hardest tasks.
The AI Industry Asked for Brakes. The President Called It a Conspiracy.
What happened: On Saturday, Anthropic CEO Dario Amodei published an essay called We Must Pace the Frontier, arguing AI companies need to deliberately slow how fast they improve model capabilities. His core worry is recursive self-improvement, AI helping build the next AI, which he says could let a swarm of AI agents become capable of taking over parts of the internet within 6 to 12 months. OpenAI's Sam Altman and xAI's Elon Musk both said they agreed within hours, and Anthropic committed to giving outside safety evaluators the same building access as employees. On Monday, President Trump rejected the idea, posting that there's a "SICK conspiracy" against AI and data centers and mocking Amodei for "pretending to be a 'perfect little angel.'"
Why it matters to you: This isn't just industry drama you can skip past. When the company that builds the AI tools you might use tells you, in public, that it's worried about how fast its own product is moving, that's worth remembering the next time a vendor pitches the newest model as risk-free. It's also a sign that federal rules aren't coming to slow any of this down. The White House has made its position clear, so any real limits on how AI gets built and sold are going to come from individual states, not Washington.
What to do about it: Nothing changes for your business today, but keep an eye on state-level AI rules, because that's where the actual restrictions are landing. See the next story.
California Just Put a Price Tag on AI Chatbots That Harm Kids
What happened: On September 10, Governor Gavin Newsom signed two laws. One bans social media platforms from using addictive features, like autoplay and algorithm-driven feeds, for users under 16. The other, called Adam's Law, requires companion AI chatbot developers to run safety risk assessments before launch, submit to independent audits, and build in default protections like usage limits and disabled notifications. Violations tied to harming a child can carry penalties up to $1 million per child, according to TechSpot.
Why it matters to you: If you or your team have been experimenting with an AI chatbot for customer support, coaching, or anything that talks with people in an ongoing, personal way, California is the state to watch. It tends to set the pattern other states copy once a tech law like this is on the books.
What to do about it: If your chatbot does more than answer basic questions, especially anything that mimics an ongoing relationship or ongoing conversation, get a lawyer's eyes on whether it counts as a "companion" chatbot under this law before you scale it up.
Apple Finally Shipped the Siri It Promised Three Years Ago
What happened: iOS 27 started rolling out on Monday, and the headline feature is a rebuilt Siri that works more like a full AI assistant than the old voice command tool. Apple first promised this version of Siri back in 2024 for iOS 18, then pulled it because it wasn't reliable enough. The full experience needs an iPhone 15 Pro or newer, launches in English only, and comes with a daily usage limit unless you pay for more, according to MacRumors.
Why it matters to you: If your business runs on an iPhone, and most small businesses do, this update is worth installing. Just don't expect it to replace ChatGPT or Claude for real work yet. It's built for everyday tasks like scheduling and quick lookups, not the research or writing your business actually needs done.
OpenAI Just Told New Customers: Not Right Now
What happened: OpenAI temporarily stopped new sign-ups for its $200-a-month ChatGPT Pro plan, saying demand for its new Astra model is straining its systems. A product lead said Pro puts the most pressure on OpenAI's infrastructure, while cheaper plans like Plus, Go, and the API remain open. OpenAI hasn't said how long the pause will last, according to TechCrunch.
Why it matters to you: If you were about to upgrade a team member to OpenAI's top-tier plan for its most capable model, you may not be able to right now. It's also a reminder that even a company as large as OpenAI can get caught flat-footed by its own product's popularity, so it's worth having a backup plan if a tool your business depends on suddenly limits access.
Meta Cut Its Managers. Now It Wants Some of Them Back.
What happened: Meta laid off roughly 8,000 employees in May and eliminated thousands of management roles as part of a push to flatten its org chart around AI. Now the company is quietly asking individual contributors inside its Applied AI division whether they'd like to move back into management. Meta made a similar move in 2023, when it pushed managers into individual contributor roles, then reversed course, according to Yahoo Finance.
Why it matters to you: AI hype has convinced a lot of companies, including some giant ones, that they could cut management layers and let AI fill the gap. Meta's own reversal is evidence that coordinating people, even in an AI-heavy operation, still needs actual humans managing actual humans. If anyone's pitched you on an "AI replaces your management layer" plan, Meta just ran that experiment and is walking it back.
Superhuman Just Bought the AI Notetaker With 400,000 Users
What happened: Email app Superhuman acquired Fathom, the AI meeting notetaker used by more than 400,000 people a month, folding it into Superhuman's existing suite of email, calendar, and AI tools. Fathom's CEO said the deal gives the smaller company access to Superhuman's 40 million users instead of trying to grow alone, according to TechCrunch.
Why it matters to you: The AI notetaker space, Fathom, Granola, Read AI, Otter, and others, is consolidating fast. If you rely on one of these tools to record and summarize your meetings, don't assume the standalone app you picked will still be standalone, or even still exist, a year from now.
Microsoft Just Let You Swap ChatGPT for Elon Musk's AI Inside Word
What happened: Microsoft is rolling out the option to use xAI's Grok models inside Copilot in Word, Excel, and PowerPoint, giving users a choice beyond the default OpenAI model. The feature is off by default and has to be turned on by an administrator, and it isn't available yet to Frontier customers in the EU, UK, or EFTA countries, according to Microsoft.
Why it matters to you: If Copilot's answers inside your Office apps have felt flat or generic, this is worth testing once it's available to you. Different models write and reason differently, and the fix for a disappointing AI answer is sometimes a different model, not a better prompt.
A Robotaxi With No Steering Wheel Just Showed Up in San Francisco
What happened: Amazon-owned Zoox launched free robotaxi rides in San Francisco with about 100 vehicles, a small fraction of Waymo's roughly 4,000-vehicle fleet in the same city. Zoox's pods have no steering wheel, pedals, or driver's seat, and passengers sit facing each other. The company is still waiting on state approval before it can charge for rides, according to TechSpot.
Why it matters to you: This one's more interesting to watch than urgent for your business, but it's a preview of how fast autonomous vehicle competition is heating up in major cities. If your business depends on local delivery or transportation costs, more competition in robotaxis and autonomous delivery is generally good news for prices down the road.
The Bottom Line
Here's the pattern this week. The people building AI are telling you, out loud, that they're worried about how fast it's moving. The government is telling you not to worry about any of it. And the actual rules that will affect your business are getting written state by state, not in Washington. You don't need to panic about any of this, and you don't need to ignore it either. You just need to keep your eyes open, because the rollercoaster isn't slowing down just because some of the people who built it are asking it to.
Enjoying the Ride?
If this issue was useful, forward it to one other business owner who's trying to make sense of all this. And if someone forwarded this to you, welcome, you can subscribe to get tomorrow's issue straight to 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