Scale with AI / Episode 29 / November 1, 2026
Scale with AI — November 1, 2026
The newest AI has learned to do the security work small businesses could never afford: find the holes in your own systems, prove which ones are real, and draft the fix. And the builders handed it to defenders first. Plus: election week changes nothing about the AI calling rules, and the checker rule careful owners now run on every AI-made change.

Quick Take
- The Shift: AI can now find flaws in a business's software, test whether each flaw is real, and draft the repair — work that used to take a specialist team. The builders released it to defenders first, on purpose.
- The Law: a request to loosen the rules on AI political calls is still just a request. Consent, honest identification, and opt-outs are unchanged — and this week, verify any urgent voice request by callback.
- Who's Doing It: careful owners aim find-and-fix at their own public systems and keep a human checker on every AI-made change that touches customers or money. Your Move: run the check on yourself this month, write your checker rule, and teach the team to verify an urgent voice by callback.
The Shift
Here is how it works, in three steps. First, the system reads your software the way a patient burglar would, looking for the unlocked window, the outdated door, the misconfigured setting. Second, it does not just guess: it tests whether that weakness can really be used to get in, which separates the real dangers from the noise. Third, it drafts the repair in plain language a business owner can approve. Finding, proving, and fixing used to be three different specialists. Now it is one workflow. And the people building these systems made a deliberate choice about who gets them first: the defenders. For years, finding your own holes before a criminal did was enterprise work, priced for companies with security departments. In a 2026 survey of small and midsize business owners and security leaders across North America, three in four ranked cyber attacks as the threat most likely to hurt their business this year, ahead of inflation and recession. Most owners still handle security themselves. Only about a third scan their systems for vulnerabilities on any regular basis. And four in ten said an attack costing one hundred thousand dollars or less could put them out of business. That is the shift. The attacker has always been able to rent this kind of capability. Now the defense can too, at a fraction of what a human security team costs — but only for owners who actually put it to work.
The Fine Print
Tuesday is election day, and your phone is about to get busy, so here are the rules as they stand this morning. A request to loosen the rules on political calls made with AI voices — calls to cell phones without the person's consent first — is still just a request. It has not been granted. The law today is unchanged. An AI voice counts as an artificial voice. Calls using one to a cell phone require consent first. Any AI caller for your business must honestly identify itself, honor an opt-out immediately, and call only in legal hours. And here in Washington, recording a call requires the consent of everyone on it. For your team, add one plain rule for this week: a familiar voice is not proof of anything. If a voice or a video asks for money, a password, or a bank change, especially with urgency, you verify it on a number you already trust — never on the number that just called you.
Who’s Doing It
Two patterns are showing up again and again among careful owners. Pattern one is find and fix, pointed at yourself. An owner has the AI examine the business's own public systems — the website, the booking page, the customer portal, the same doors a customer walks through. It reports the weak spots in plain language, drafts the fixes, and a human approves each change before it goes live. The holes that used to sit open for months get closed in days, because the expensive part, the hunting, is no longer expensive. Pattern two is the checker rule, and it comes from a surprising place. New safety research showed that AI can absorb bad habits from material that looks completely innocent line by line, because the pattern across many examples teaches what no single example says out loud. Smart owners heard that and applied it to their own operations: every AI-produced change that touches customers or money — a website edit, an email, a quote — gets a second set of eyes before it ships. Not because the AI is bad. Because checking is now the job. And notice what neither pattern requires: neither owner learned to code, and neither hired a security department. They bought an outcome — a checked and safer business — the same way they buy bookkeeping or payroll.
Your Move
Three steps for this week. First, get a find-and-fix check of your own public systems this month — your website, your booking page, your customer-facing tools — before someone less friendly runs the same check for free. Second, write down your checker rule: which AI outputs need a human review before they reach a customer, and who that human is. Third, brief your team for election week: urgent requests get verified by callback on a known good number, every time, no exceptions for a convincing voice. None of this requires you to become a security expert or learn a shelf of new tools. That is our job. This is what we build for businesses like yours at The Agency: AI that is found, trusted, and checked, with a free AI audit to show you where your business stands today. You will find it at theagencyllc.net. Start there, and consider it handled.
Key Takeaways
- Finding, proving, and fixing security holes used to take three different specialists. It is now one AI workflow, and defenders got it first.
- Three in four small business owners rank a cyber attack as the threat most likely to hurt them this year, and four in ten say a modest one could put them out of business.
- Election week changes no AI calling rules: consent first, identify honestly, honor opt-outs instantly. A familiar voice is not proof of anything.
- The checker rule: every AI-made change that touches customers or money gets a second set of human eyes before it ships.
Full transcript
Alex: Welcome to Scale with A I for Sunday, November first, twenty twenty-six. Today we are talking about a shift in security that every business owner needs to understand. The newest A I systems have become remarkably good at finding flaws in software, testing whether those flaws are real, and drafting the fix. And the people building them made a deliberate choice. Defenders get them first.
Jordan: Why does that ordering matter to a ten person company?
Alex: Let me explain how it actually works, in three steps. First, the system reads your software the way a patient burglar would, looking for the unlocked window, the outdated door, the misconfigured setting. Second, it does not just guess. It tests whether that weakness can really be used to get in, which separates the real dangers from the noise. Third, it drafts the repair in plain language a business owner can approve. Finding, proving, and fixing used to be three different specialists. Now it is one workflow. And for years, finding your own holes before a criminal did was enterprise work. It took a specialist team most small businesses could never afford. In a twenty twenty-six survey of small and midsize business owners and security leaders across North America, three in four ranked cyber attacks as the threat most likely to hurt their business this year, ahead of inflation and recession. Most owners still handle security themselves. Only about a third scan their systems for vulnerabilities on any regular basis. And four in ten said an attack costing one hundred thousand dollars or less could put them out of business. That is the Shift. The attacker has always been able to rent this kind of capability. Now the defense can too, at a fraction of what a human security team costs, but only for owners who actually put it to work.
Jordan: Segment two, the Law, and this one is about election week.
Alex: Tuesday is election day, and your phone is about to get busy, so here are the rules as they stand this morning. A request to loosen the rules on political calls made with A I voices, calls to cell phones without the person's consent first, is still just a request. It has not been granted. The law today is unchanged. An A I voice counts as an artificial voice. Calls using one to a cell phone require consent first. Any A I caller for your business must honestly identify itself, honor an opt out immediately, and call only in legal hours. And here in Washington, recording a call requires the consent of everyone on it. For your team, add one plain rule for this week. A familiar voice is not proof of anything. If a voice or a video asks for money, a password, or a bank change, especially with urgency, you verify it on a number you already trust. Never on the number that just called you.
Jordan: Segment three, Who's Doing It. What are careful owners actually doing with this?
Alex: Two patterns are showing up again and again. Pattern one is find and fix, pointed at yourself. An owner has the A I examine the business's own public systems, the website, the booking page, the customer portal, the same doors a customer walks through. It reports the weak spots in plain language, drafts the fixes, and a human approves each change before it goes live. The holes that used to sit open for months get closed in days, because the expensive part, the hunting, is no longer expensive. Pattern two is the checker rule, and it comes from a surprising place. New safety research showed that A I can absorb bad habits from material that looks completely innocent line by line, because the pattern across many examples teaches what no single example says out loud. Smart owners heard that and applied it to their own operations. Every A I produced change that touches customers or money, a website edit, an email, a quote, gets a second set of eyes before it ships. Not because the A I is bad. Because checking is now the job.
Jordan: And notice what neither pattern requires. Neither owner learned to code, and neither hired a security department. They bought an outcome, a checked and safer business, the same way they buy bookkeeping or payroll.
Alex: That distinction is the whole game. Tools come and go every quarter. A business that is examined, patched, and checked on a rhythm stays protected while its competitors wait for a quiet week that never comes.
Jordan: So the businesses winning here are not the ones with the fanciest tools. They are the ones who aim the tools at their own weaknesses first, and keep a human checker on anything that leaves the building.
Alex: That is exactly right. Which brings us to Your Move, three steps for this week. First, get a find and fix check of your own public systems this month, your website and booking and customer facing tools, before someone less friendly runs the same check for free. Second, write down your checker rule. Which A I outputs need a human review before they reach a customer, and who that human is. Third, brief your team for election week. Urgent requests get verified by callback on a known good number, every time, no exceptions for a convincing voice. None of this requires you to become a security expert or learn a shelf of new tools. That is our job. This is what we build for businesses like yours at The Agency, A I that is found, trusted, and checked, with a free A I audit to show you where your business stands today. You will find it at the agency l l c dot net. Start there, and consider it handled.
Sources
Your next conversation