It has long been my contention that the AI buildout—which unquestionably is the dominant force underlying the current market environment—will ensure abundant liquidity for equities for as long as it persists.
It can further be argued that the AI investment theme is today’s analog to the computer tech buildout of the 1980s or the Internet stock boom of the ‘90s and, as such, is doing much of the heavy lifting for the major indexes.
For the last couple of years, investors who were awake to this reality were able to operate more or less on cruise control, secure in the knowledge that AI-related liquidity would almost certainly support the tech sector through whatever geopolitical or economic obstacle might emerge along the way, specifically, the fact that the AI buildout was in its early innings and presumably had a long way to go before completion.
[text_ad]
This point cannot be understated, for it calls to mind the similar boom in tech stocks that dominated the mid-to-late 1990s during the Internet infrastructure buildout phase. In similar fashion, as long as the Internet buildout was in force, the Nasdaq had a solid backstop.
However, when the Internet buildout was largely complete (or was at least significantly advanced), that’s when the Internet stock boom ran into trouble. In a similar vein, I think we can assume that the current AI buildout will follow much the same path. That is, as long as the data center infrastructural development remains largely incomplete, stocks in the aggregate are likely to remain buoyant as the massive amounts of money flowing into AI-related themes should support the broader market.
The good news through much of this year is that it was universally assumed that AI development was still in its early stages. But it now appears the buildout has entered its intermediate phase. And while it still suggests liquidity should remain ample enough going forward, since the buildout isn’t yet in its advanced stages, it means the heady days of investors operating on cruise control are likely numbered. To wit, the low volatility environment that has persisted for much of this year is likely to give way to more volatility heading into 2027.
Again, this isn’t to say that a bear market is imminent. I maintain that bullish conditions, overall, are likely to persist well into next year. But if the AI buildout is indeed entering its middle stages, we can look to the computer tech and Internet stock buildout analogs of the past for clues to what we can likely expect from here.
Before discussing that, I need to point out my reasoning for assuming the AI buildout is entering its middle stage. The recent appearance of fear surrounding the pace of AI development is, I think, a major piece of evidence. Underscoring this concern was a post on the social media platform X that made news headlines when former Anthropic and OpenAI researcher Jason Coxon warned that AI labs are racing toward self-improving superintelligence and gambling with human lives.
His post sent shockwaves through the AI industry, as well as among investors, especially as its implication was that increased government regulation would be a likely next step. On that score, several leading AI companies and executives are now asking the federal government for more regulation or oversight of advanced AI. And that, in turn, can be taken as a sign that the big industry players are looking to weed out their smaller competitors as the AI buildout advances.
The AI Buildout vs. the Internet Buildout
This is not unlike what we saw during the ‘90s internet buildout boom when “wildcat” independent internet startups were nibbling at the heels of the bigger companies during the early phase of the boom, and in doing so, eating into their bigger competitors’ margins while also increasing concerns among federal regulators over internet security and consumer protection. The end result was that by the latter part of that decade, many of the smaller players were weeded out by the increased regulations that followed.
In a similar vein, I think a case can be made that then-Fed Chairman Alan Greenspan’s infamous “irrational exuberance” speech in December 1996 marked the start of the middle phase of the internet infrastructural buildout. And indeed, it was roughly from that time onward that tech sector volatility began to steadily increase as the buildout became more advanced and liquidity became less profuse while speculative excesses also became more pronounced.
Granted, there is a danger in extrapolating this analogy too far, particularly as concerns over the pace and risks of the AI buildout intensify. Yet the regulatory environment is also beginning to change: OpenAI, for example, has recently called for mandatory, capability-based federal AI safety requirements, including testing, independent assessments, cybersecurity protections and incident reporting. If such regulation gains momentum, it could eventually favor larger, better-capitalized players and contribute to consolidation within portions of the AI ecosystem.
Exploring the Next Stage of the AI Buildout
That in turn wouldn’t necessarily mean that AI infrastructure spending immediately declines; rather, the next phase could involve a gradual shift from rapid expansion and proliferation toward greater concentration, tighter capital discipline and a smaller number of dominant players. And that would almost certainly mean shrinking liquidity for capital markets as the final stages of the AI buildout approach.
I view this development (which almost certainly has a long way to go before it completely plays out) as being a double-edged sword for turnaround investors. On the one hand, the early stages of the AI buildout haven’t been entirely kind to turnaround stocks in certain sectors and industries—the exception, of course, being AI and infrastructure-related companies undergoing turnarounds, e.g., Intel (INTC).
But companies experiencing a financial turnaround that aren’t directly tech-facing have admittedly had a much tougher go of it this year. Investors seem content to focus their attention on AI and related spaces while largely ignoring “old economy” sectors, resulting in a great many turnaround stocks languishing as investment funds are allocated elsewhere.
Why the Middle Stages of the AI Buildout Should Favor Turnaround Investors
I believe that paradigm could change, however, as the middle phase of the AI buildout becomes more apparent to investors heading into 2027. And as tech sector volatility presumably increases, investors will be incentivized to turn their attention to the value-oriented aspects of the market, which in turn should benefit the largely ignored (non-tech-related) turnarounds.
That prediction would align with what happened starting in 1997 when the Internet buildout entered its middle phase and a sector rotation into utilities, biopharma and retail stocks began as a cooling of internet “mania” among investors was in evidence. By 1998, moreover, investors were increasingly sensitive to valuation and were actively hunting for bargains across a large number of non-tech sectors.
With that in mind, I think turnaround and deep-value investors can take heart: Patience is apt to be rewarded at some point during the second half of the AI buildout phase, which, by my reckoning, is now upon us.
[author_ad]
