Erich Schmidt LinkedIn

Boise, Idaho

Today’s brief Read

01Profile

“The tool said so” is never a good enough answer.

I learned that while working digital forensics cases in law enforcement. It still shapes how I build, teach, and use technology, including AI.

Connect on LinkedIn Read today’s brief
Erich Schmidt, smiling, in a dark blazer and white shirt
Erich Schmidt · Boise, Idaho

02Today’s brief

Here’s what broke overnight.

  1. Each brief covers what broke, why it matters to a small business, and what to do about it. Every finding links to its source.

An AI model drafts this brief from public data, and it’s published automatically. No one reviews it first, so check the linked source before you act. Here’s how it’s made.

Read the full brief RSS

03What I believe

  1. Judgment comes before tools.

    Most AI projects fail at the question, not the model. Decide what’s worth doing before you decide what to do it with.

  2. Trust is earned.

    AI should get access the way a new hire does: a little at a time, with someone checking the work.

  3. Train the people, not just the prompts.

    The skill that matters most is knowing when the output is wrong and having the confidence to say so.

  4. Proof beats hype.

    If you can’t measure whether it works, you can’t defend it, keep it, or scale it.

  5. Every automation is an attack surface.

    Build it as if someone will try to break it, because someone will.

04Built in public

Open source

SMB + AI Threat Feed

This is the engine behind the brief. It pulls from free public sources, ranks each flaw by how likely it is to be exploited, and turns the results into clear next steps for small businesses.

View on GitHub

Open source

Möbius

Write the rules for an AI agent before anyone builds it. Möbius is a local command-line tool that turns a rough idea into a written spec, a risk check, and a twelve-question intake, and then it stops.

View on GitHub

05About

I started as an investigator, and I still work like one.

Law enforcement came first. I worked digital forensics cases, where a wrong call has real consequences.

Since then, I’ve trained thousands of people in forensics, incident response, cybersecurity, and OSINT. Along the way, I’ve published research, disclosed vulnerabilities, and helped protect senior executives from online attacks.

That’s the lens I bring to AI. I don’t treat it as a novelty. I treat it like any other system: someone checks its work, someone thinks about how it could be attacked, and someone owns the consequences. The technology is moving fast, but the judgment to use it well is scarce, and it can be taught.

I live in Boise, and I’m active in the Treasure Valley cyber community, where I mentor, volunteer, and host local meetups. I also run Arrowrock.AI, which helps local small businesses fix their processes, automate routine work, and use AI where it helps.

06Contact

Let’s talk.

For speaking, training, mentoring, or a conversation about security and AI, LinkedIn is the best way to reach me.

Connect on LinkedIn GitHub