Investments, Artificial Intelligence and the OBBA

Christopher D. Flis | Oct 9, 2025

Investments

2025 is shaping up to be a very nice year for marketable securities. While tariffs seemed to upset the apple cart in April, markets have mostly shrugged off their impact and continue the march upward. Remember, market downturns are a "when", not an "if", so please adjust your expectations accordingly.

While there is still another quarter to go, International Markets have proven the place to be so far this year. I owe nothing to my predictive abilities; however, I do owe much to the importance of diversification and perseverance. Diversification GUARANTEES that a portfolio manager will have to "apologize" for the performance of part(s) of an investment portfolio frequently....that's the dim side.

On the bright side, diversification also acknowledges the unpredictability of markets by providing exposure to areas of capital markets to which a singularly focused portfolio does not. For most US investors, International Markets are either completely absent or severely under-represented in their portfolios.

At RAMCO, we advise healthy exposure to markets outside the US. While capitalism's crown jewel is the United States, other countries on our globe are also riding the capitalist wave - and from time to time they leave the US in the dust...2025 is one of those years. Over time, the horse race is pretty even'ish.

Artificial Intelligence (Impact and Investment)

So all of you know, RAMCO is WELL aware of Artificial Intelligence. We are also equally aware of the format most AI Agents take - prompt to output, iterate, etc. And we do appreciate when we are forwarded the output of any AI Agent to, for example, "see what Grok said". Indeed, the Agent outputs have become quite impressive.

So you also know, we are frequent users of the same AI agents, though we do not use any SPECIFIC Client Personal Data in any AI Agent - for reasons we will explain in a bit.

From a societal perspective, I think it's fair to say that the range of AI's impact ranges from "Life Changing" to "Negligible". I would caution against placing faith in ANY prediction indicating certainty. The technology is just too new to know exactly how things are going to unfold. So I won't make any predictions about where we are going. Though it is safe to say that the genie is out of the bottle and AI is creeping into many crevices of our lives.

From an Investment Perspective, my suggestion is to keep a wary eye on where AI-related investments are placed. The most recent example we have of "life changing" technology was the internet itself back in the late 1990s. I am old enough to remember both the mania around the "Dot Com" bubble and some of the carnage it left in its wake.

A specific example is JDS Uniphase (Ticker: JDSU). This was an "optical networking" company that was a darling of the latter stage of the internet mania. At its peak, JDSU reached a peak market capitalization of $125 Billion ($1,170 per share) in July of 2000 - $125 Billion was real money back then....not the pocket change it is now. By late 2008, sadly, you could buy JDSU shares for $2 each - a 99.8% loss...ugh. (SOURCE)

Ultimately, JDSU split into two companies:

VIAVI Solutions: Current Market Cap $2.845 Billion

Lumentum: Current Market Cap $11.936 Billion

I tell you this not to alarm you. I simply want to alert you to the fact that rapid technological change and outsized market returns sometimes don't mix well....even though intermediate market valuations may feel quite good. A better explanation from the 1986 Berkshire Hathaway Annual Letter is as follows:

"Severe change and exceptional returns usually don't mix. Most investors, of course, behave as if just the opposite were true. That is, they usually confer the highest price-earnings ratios on exotic-sounding businesses that hold out the promise of feverish change. That prospect lets investors fantasize about future profitability rather than face today's business realities. For such investor-dreamers, any blind date is preferable to one with the girl next door, no matter how desirable she may be." - 1986 Berkshire Hathaway Annual Letter

One aspect of AI's impact frequently reported is just how intense the capital investments are for the data centers housing all this "compute". What's less well-reported is that revenue for one company MUST be an expense for another company. Expenses are financed by either cash, debt, or revenue. Here is an example using Open AI (A non-publicly traded company with which I'm sure many of you are familiar):

OpenAI recently purchased $300 Billion in computing power over the next 5 years from Oracle
Assuming this purchase is equally spread over 5 years, that is $60 Billion per year - AI can't bend mathematics just yet regardless of what Grok says
Further assuming that Open AI intends to make a 25% profit (return) on it's investment, revenue from just this investment needs to be $75 Billion Annually
For the first 6 months of 2025, Open AI's revenue was $4.3 Billion (SOURCE)
- I don't care what company you are or who you are that's leading it, 10X'ing revenue is a true Herculean endeavor
- As a comparison, Alphabet (Google) - one of the greatest growth companies the world has EVER seen with a near monopoly on search - had revenue in 2024 of $350 Billion
- In 2011, Google's revenue was about $37 Billion, or 1/10 of 2024's
- Stated differently, Google needed 14 years for its most recent 10X
I may have over-simplified Open AI's business somewhat and may be missing some super-profitable aspect of their business model - however, sooner or later the reality of Balance Sheets and Income Statements come to ALL companies
And if there is ever a rationalization, the valuation corrections can be JDSU-like
That's the potentially bad news

The good news is that by being invested in the S&P 500 with a portion of your portfolio, you are "in" on the publicly-traded companies most leveraged to AI's outcome - think Meta, Alphabet, Amazon, Nvidia, etc. Even better, the "old" economy stocks are included in the index as well, so if AI does have a reality check, other companies can compensate.

Speaking of an "old" economy stock, let's look at the trusty Coca-Cola Company. In 1999, I'm pretty sure that when asked to compare the 25-Year performance expectation of Coke vs. JDSU, you'd have found more than a few who would have said Coke would be left in the dust. Let's look at the scorecard (I'm listing share prices because Market Cap is a distorted number due to the mountain of share buybacks Coca-Cola has done in the past quarter century):

The Coca Cola Company Share Price on 12/31/1999 (Split Adjusted): $29.13

The Coca Cola Company Share Price on 9/30/2025 (Split Adjusted): $66.32

Total Dividend Distributions per share (2000 - 2025): $28.92

Not a bonanza, though certainly better than JDSU's Negative 99.8%

Maybe AI will replace the smile induced when one drinks a Coke - I seriously doubt it...I think it's a safe bet that in the future we'll be producing and consuming Coca-Cola products on Mars.

Artificial Intelligence (For use in Financial Planning)

You may perhaps be curious about AI's use in RAMCO's services. First, some background on compliance. For every technology solution we use, the Financial Services Industry Security Standard is SOC 2.0. The Enterprise (Pay) Versions of most of the more popular AI Agents are indeed SOC 2.0 compliant....that's good. Happily, there are also AI services to which one can subscribe that provide a SOC 2.0 "walled garden" within which one can use the various AI Services like ChatGPT, Grok, Gemini, Co-Pilot, etc.

Assuming Security is sorted - a pretty safe bet - then we move on to what the current crop of AI Agents produce. The results are somewhat mixed. On one hand, ChatGPT can pass the CFA Exam. On the other hand, AI Agents fare poorly when answering Tax-Related Questions. And therein lies the problem with AI right now - attribution. Just as self-driving cars do not GUARANTEE the absence of auto-related fatalities, AI Agents do not GUARANTEE their results on tax topics, or any topic. And even worse, AI can hallucinate and provide glowing, neatly formatted output that is precisely wrong. And for the professional who relies upon the output, "the dog ate my homework" response is not going to cut it.

Here are a few examples of what financial-related errors I've seen from AI Agents:

Mistaking the Qualified Business Income Deduction (QBI) as 23% when it is 20%
Not using Itemized Deductions on a tax Estimate when the Standard Deduction is exceeded
Severely underestimating the cost of Homeowner's Insurance
Not recognizing that Social Security must be claimed at Age 70 as there is no longer a benefit to waiting beyond that age to claim
And if you are looking for predictive ability about where capital markets are headed - be very careful....that is an unlikely benefit of a singular AI Prompt.
Anecdotally, I asked Grok about the Arkansas-Memphis football game a few weeks ago: Grok predicted a Razorback walkover....Arkansas LOST 32-31 and they have a sort of new Head Coach now. Grok is 0-1 in picking football games in my book.

I fully realize that every single thing I have said about AI can be refuted - and I may look very foolish once the realities of AI unfold. Right now, as the Steward of what is now about $70 Million of investable assets at Resilient Asset Management, I am telling you that my policy is to use AI as an able-assistant and no more. No recommendation or conclusion we present to a Client will be solely AI-generated. Of course, AI capabilities will evolve over time and my policies will adjust accordingly. For now, you're stuck with me arbitrating over the recommendations we provide to you.

The One Big Beautiful Bill Act (OBBA)

As we enter the final quarter of the year, Income Tax motivated Financial Planning becomes a bit more pressing. For those with more complex tax situations, our tax dialogue with you is a bit more rhythmic. For others, now is a good time to check on your 401(k) contributions, IRA Contributions, and other matters (RMDs, Charitable Giving, etc.) prior to year-end so any issues can be addressed prior to 2026. We, of course, are at the ready to assist as needed.

On another note, the One Big Beautiful Bill Act (OBBA) - in case anyone forgot - was passed this year. It was a gargantuan piece of legislation just about on-par with the Tax Cuts and Jobs Act in 2017. Because of the many changes to the tax code, we have prepared a series of videos for your review. We broke things down into 3 sessions - limited AI was used in editing, but not in the recording...that's me talking, for better or worse. The three sessions are as follows with links to the videos provided:

Introduction to OBBA

Small Business and Solopreneur Considerations

Charitable Giving

The entire series is about 32 minutes in total...I can't promise it's as gripping as Severance, though there are some important tidbits in there that apply to all of you.

Conclusion

I realize this was a lengthy letter - I hope you found it worthwhile. As always, if you would like to schedule time to discuss your personal situation, please contact us directly so we can get you on the calendar.

Thank you for your continued trust in Resilient Asset Management in helping you achieve your financial goals.