By Vaughn Woods, CFP®, MBA
There is a quality in today’s most impressive AI systems that I have come to think of as flatness. Not shallowness — the models are not shallow. They can summarize a 10-K, translate a shareholder letter, and produce a serviceable market commentary before my coffee cools. What they cannot do, and what I did not fully appreciate until I began working with them daily, is register the difference between a market that is merely repeating itself and a market that has quietly begun to mean something new.
That difference is the entire job.
What flatness looks like in practice
A transformer processes context as a widening statistical record. Every new token is weighed against every earlier token, and the model returns the most probable continuation. This is a remarkable achievement, and it is why the fluency feels so uncanny. But probability is not salience. The system has no mechanism by which one sentence in the middle of a filing suddenly matters more than the two hundred sentences around it. Nothing tightens. Nothing leans in. The signal and the noise arrive with the same posture.
A seasoned portfolio manager does the opposite. Somewhere in the second paragraph of a conference-call transcript, a CFO uses a verb she has never used before, and an old part of the brain — the same circuitry that once kept our ancestors alive by noticing which rustle in the grass was not the wind — quietly reallocates attention. The manager does not always know why the sentence matters. She only knows that it does, and that the knowing came before the reasoning.
I wrote my master’s thesis in 2007 on this problem. I called it a neuroeconomic solution set for the integration of intelligence models, and its central claim was that meaning in the human brain is built inside a structure I described as a psycho-hyper-vector-space — a living, context-sensitive geometry in which prediction error, emotional salience, and prior experience continuously reshape which inputs count. Two decades later the machines have caught up to the vector-space metaphor. They have not caught up to the living part.
Why this matters for capital
Markets do not punish investors for missing patterns. They punish investors for confusing the familiar pattern with the truly new. Every cycle I have worked through — 2000, 2008, 2020, the concentrated drawdowns of the last two years — began with a period in which the surface data looked ordinary and something underneath had shifted. The advisors who protected client capital were not the ones with the most information. They were the ones whose attention tightened before the story became legible.
An AI cannot tighten. It can only widen. Ask it whether this quarter’s guidance language resembles prior quarters and it will tell you, accurately, that it does. It cannot tell you that the resemblance is itself the tell — that management has begun to sound rehearsed in a way it did not sound eighteen months ago. That judgment requires a felt sense of deviation from a baseline the model does not possess, because the model is never bored, never surprised, and never afraid.
Novelty, in the human brain, is how we escape boredom. It is the reason a portfolio manager who has read forty earnings calls this season will feel the forty-first one differently if something is off. The nervous system rewards the detection of genuine change. A language model has no such reward. It treats the forty-first call as one more sample in a distribution, and its confidence rises with familiarity rather than falling.
Figure 1. Widening context vs. tightening attention.
What I use AI for, and what I do not
I use these tools daily. They draft, they summarize, they cross-reference, they surface citations I would have taken an hour to find. For anything that benefits from breadth without judgment, they are the best assistants I have ever had. I recommend them to clients for the same reason.
What I do not delegate to them is the moment of interpretation. When a client asks whether a sector concentration is beginning to look like the last one that hurt us, or whether a piece of policy language is a signal or a shrug, that question does not belong to a probability distribution. It belongs to a person whose attention has been shaped by thirty years of watching markets behave, misbehave, and occasionally lie. The AI can prepare the ground. It cannot make the call.
This is not a defense of advisors against machines. It is a description of a division of labor that the mathematics of transformer architectures makes permanent for the foreseeable future. Until a model can experience prediction error as something other than a loss function — until it can be surprised — its confidence and its accuracy will drift apart in exactly the conditions where clients most need them to converge.
The fiduciary implication
A fiduciary standard is, at its core, a promise about attention. It says that when your circumstances change in a way you may not yet have noticed, someone is watching who will notice on your behalf. That promise cannot be automated, because the noticing is the part that requires a nervous system with skin in the game. Fee transparency, tax coordination, and disclosure discipline are the visible parts of the standard. The invisible part is the tightening of attention when something in your life, or in the market, begins to mean something new.
The seasoned advisor’s edge, in an age of impressive and flat machines, is not that she knows more. It is that she can still be surprised, and that her surprise arrives early enough to matter. That is the work. It always has been. The tools have simply made the shape of the work easier to see.
References
Woods, V. L. (2007). Context: A Neuroeconomic Solution Set For The Integration of Intelligence Models. Master’s thesis, Point Loma Nazarene University.
Disclosures
Vaughn Woods, CFP®, MBA is President and Founder of Vaughn Woods Financial Group, Inc., an Investment Advisor Representative of Bolton Global Capital, Inc. Client assets are held in custody through Pershing LLC, a subsidiary of Bank of New York Mellon. This article is for informational purposes only and does not constitute personalized investment or tax advice.
We are unable to accept orders via email. If you wish to place an order, please consult your registered representative or contact the home office trading desk at (800) 649-4554.
This email system is for business purposes only and any information, including attachments, transmitted in this email is not confidential. Any message may be reviewed by authorized compliance personnel and/or produced to regulatory agencies or others with a legal right to access such information.
Past investment performance is not indicative of future results. Securities offered through Bolton Global Capital, Inc., Bolton, MA. Member FINRA, SIPC. Advisory services offered through Bolton Global Asset Management, a registered investment advisor, 579 Main St., Bolton, MA 01740 (978) 779-5361.
Investors should be aware that there are risks inherent in all investments such as fluctuations in investment principal. Past performance is not a guarantee of future results. Asset allocation cannot assure a profit nor protect against loss. Although the information has been gathered from sources believed to be reliable, it cannot be guaranteed. Views expressed in this newsletter are those of Vaughn Woods and Vaughn Woods Financial Group and may not reflect the views of Bolton Global Capital or Bolton Global Asset Management. The information provided is for general informational purposes only and should not be considered individual recommendation or personalized investment advice. Representatives and Advisors of Vaughn Woods Financial Group are not tax or legal professionals, if you need tax or legal advice, please make sure to consult a tax professional/CPA and/or a lawyer.VW1/VWA0405