Paul Osterberg

We Help Companies Understand and Mitigate Cyber Risk. CISO Support Services, Cybersecurity Risk Assessments, Vendor Risk Management

[Aaron @aaronspradlin]

AI in wealth management is trapped between two bad takes: either it will replace advisors tomorrow, or we should just run wild with whatever consumer AI tool is popular this week.

Both completely miss the point.

I recently joined Mustafa Baseri (Co-Founder of @hamachiai ) on the Advisor Perspectives podcast (@ThinkAdvisor) to cut through the marketing noise and talk about the conversation our industry actually needs right now: Governed adoption.

When it comes to AI, my philosophy is straightforward:
“We’re not wanting to be the firm to say no, but we’re also not wanting to be the firm to create the illusion of yes.”
Banning AI is not realistic. But telling advisors to plug sensitive client data, CRM records, or financial plans into unapproved consumer tools isn’t leadership either—it’s a major compliance liability.

Real AI adoption in financial services requires enterprise guardrails: audit trails, archiving, strict permissioning, source control, and human oversight.

Here are the 3 core signals from our conversation:

“Free is Not Free”: In a regulated industry, cheap or unmonitored AI tools create hidden costs like data exposure, weak supervision, and poor recordkeeping.

The “Golden Source of Truth”: Wealth management AI shouldn't rely on random internet scraping. True efficiency happens when AI is securely containerized and connected to permissioned internal data: your CRM, portfolio records, and firm WSPs.

Agentic AI & Cyber Risks: We are entering an era where AI agents can read, draft, and act. But automated tools create new attack surfaces, like "no-click" email threats. A "human-in-the-loop" design is non-negotiable. AI can draft the email, but the advisor must remain the one who hits send.

The Bottom Line:
AI can assist the advisor, but it does not replace the advisor’s fiduciary responsibility. Financial advice is not just data delivery—it is judgment, context, and trust.

AI will absolutely change how we work, help us prepare better, and let us serve clients more efficiently. But in financial services, access is not the same thing as safety.

If you are trying to navigate how to safely scale your advisory practice with these technologies, check out our full discussion on ThinkAdvisor. 👉 https://t.co/RhzSwTyaJA

[Paul Osterberg @POsterberg]

@aaronspradlin Brainy 🧠 🧐

[Paul Osterberg @POsterberg]

@aaronspradlin @sharbel Reagan said, “Trust, but verify”. That won’t work for Agentic AI. We need to “Not trust, AND verify.”

-Trusted vs untrusted inputs
-Sandboxed tools/limited permissions
-Human approval for sensitive actions
-Structured APIs vs. raw browsing
-Zero-trust architectures

[Bloomberg @business]

EXCLUSIVE: Treasury Secretary Scott Bessent and Federal Reserve Chair Jerome Powell summoned Wall Street leaders to an urgent meeting on concerns that the latest AI model from Anthropic will usher in an era of greater cyber risk. https://t.co/d8eVXEzq5v

[Paul Osterberg @POsterberg]

@jakesgoodopines @SidKhurana3607 A combination of: 1) French Canadians from Quebec and 2) lots of immigrants that worked in the iron ore mines from places like Italy, Poland, Ireland, Croatia, Slovenia, etc

[Paul Osterberg @POsterberg]

@jaketapper @ProjectLincoln Ugly

[Gopherated〽️ @Gopherated]

Gophers student section was rocking all season long

[Paul Osterberg @POsterberg]

@interesting_aIl @grok who is the musician in this video

[Paul Osterberg @POsterberg]

@DrkMvrkIX @FanDuel Agreed. Klatt is excellent, and Gus is horrible.

[Paul Osterberg @POsterberg]

@FanDuel Nauseating

[Paul Osterberg @POsterberg]

God, I offer myself to Thee-to build with me and to do with me as Thou wilt. Relieve me of the bondage of self, that I may better do Thy will. Take away my difficulties, that victory over them may bear witness to those I would help of Thy Power, Thy Love, and Thy Way of life. May I do Thy will always!"
(p. 63)

God, Take my will and my life. Guide me in my recovery. Show me how to live. AMEN
(p. 59)

[Paul Osterberg @POsterberg]

@Lane_Kiffin Page 86. Great way to start every day!!!

[Paul Osterberg @POsterberg]

@aaronspradlin AI in your hands has made you an Einstein. Bring on the March experiments!

[Paul Osterberg @POsterberg]

@aaronspradlin Einstein is likely “watching” your work with great interest.

[Paul Osterberg @POsterberg]

Want to be a secure software engineer or cloud architect? We ethically hack companies every week. Learn what we know.
https://t.co/MOiYxtSipI

[Paul Osterberg @POsterberg]

@"The Hack: How a Threat Actor Exploits a Financial Institution"https://t.co/kP5fiAqLyP on ⁦@LinkedIn⁩

[Paul Osterberg @POsterberg]

https://t.co/5Kba18mUf6

[Paul Osterberg @POsterberg]

@"Application Penetration Testing: A Critical Component of Modern Cybersecurity"https://t.co/97r4zlCWuD on @LinkedIn

[Paul Osterberg @POsterberg]

@aaronspradlin “So you’re saying the universe runs on quantum teamwork, galactic déjà vu, and some cosmic-level multitasking? Makes my daily to-do list feel pretty basic. Coherence really is the ultimate life hack!”

- A cogent response from a human with a little assistance from AI