Beyond the Past and Before the Future: Interpreting the Present State of Artificial Intelligence in Private Markets
- Market commentary
September 23, 2025 | 5 min read
Managing Director,
Senior Market Strategist
A provocation worth consideration
At the Paris AI Action Summit this past February, Dario Amodei, CEO of AI superpower Anthropic, offered a striking analogy: As early as 2026 or 2027, AI systems will be “akin to an entirely new state populated by highly intelligent people appearing on the global stage—a ‘country of geniuses in a datacenter’—with the profound economic, societal, and security implications that would bring.” If this is true, I wonder if this “AI nation” will have a seat at the table when and if we need to renegotiate sovereignty, trade, governance, or climate policy? And which large language model will inform its worldview?
Flies on the walls of our investment committee meetings can hear us regularly debate the future state of AI. The tone may be less theatrical, but the questions are no less consequential:
- How do we best give our investors access to what is one of the most transformational innovation cycles in history?
- How do we contemplate business models that are sound and relevant today but may be obsolete tomorrow?
- How do we balance our fiduciary duty to generate returns with our responsibility to consider the broader societal implications of the AI transformation we are helping to fund?
Contemplating our answers to these types of questions can guide how we interpret recent AI deal activity and help us understand how its key players are positioning themselves for a vastly different future state.
Deal activity as a mirror of the present
AI today is propelled by extraordinary momentum—powered by surging funding, early consolidation, strategic partnerships, and a race to define what this massive innovation cycle will become. To understand where we are going, let us remind ourselves where we have been.
Five years ago, AI was a $50 billion curiosity. Today, adjacent sectors are approaching a $1 trillion market size, accelerated by the “ChatGPT moment” in November 2022 and catalyzed by funding from large cap incumbents as well as an intensely competitive venture investment environment and unprecedented venture investment activity.
- VC invested over $100 billion into this massive innovation cycle in 2024 alone; a figure we could see double in 2025.
- JPMorgan projects that in 2025, major hyperscalers will spend around $300 billion on AI infrastructure, a staggering leap from years prior.
- Morgan Stanley estimates generative AI could contribute up to $10 trillion in global GDP, challenging the zero-sum game theory, and driving towards potentially a $3-$5 trillion market over the coming decade(s). (Note: this is a projection—not yet realized.)
This rapid scaling raises urgent questions about concentration of AI capabilities and their broader societal distribution. I see several key factors currently influencing capital market activity.
Investors have shown an unrelenting appetite for foundational models
Market dynamics are being reshaped at unprecedented speed, with AI companies setting new records in venture financing.
OpenAI’s $40 billion private market raise earlier this year meaningfully surpassed Saudi Aramco’s record $25.6 billion public market raise from 2019, signaling a seemingly insatiable demand for foundational AI models, regardless of whether it is funded by the public or the private markets.
Anthropic has nearly tripled its valuation in a relative blink of an eye, ballooning from $61.5 billion in March to $183 billion in September following a recent $13 billion round. The company’s run-rate revenue soared from circa $1 billion to over $5 billion in just a few months.
As these foundational models become more powerful and concentrated among fewer players, questions of access and governance become increasingly pressing. Nevertheless, this momentum is generating—or even forcing—complementary strategic initiatives across the ecosystem.
Strategic dealmaking is the new race
Meta’s acquisition of a 49% stake in Scale AI underscores a strategic pivot to secure both next-gen data infrastructure and AI talent—essential ingredients for “data at scale” in generative AI leadership. It also proves if Meta’s original path of “build for the future” was giving ground to the competition, they have quickly pivoted to “buy for the future” to find the way back to a position of advantage. The current debate, though, is whether this marriage of convenience—or perhaps even necessity—will endure.
Alphabet’s proposed $32 billion acquisition of Wiz underscores the rising importance of AI-native security. As AI becomes deeply embedded, platforms that guard the trust layer become essential. Now we must watch to see if this proposed acquisition makes it past the regulators. For Wiz, with a 10% break-up fee as part of the prenup, this proposal was worth accepting the risk of running the regulatory gauntlet.
The AI enablement of incumbents through M&A
ServiceNow’s $2.8 billion acquisition of Moveworks signals how conversational AI is being woven into enterprise workflow. Siemens’ $5.1 billion purchase of Dotmatics shows AI’s transformative impact on R&D and product innovation. In both cases the catalyst was AI enablement.
Initial market skepticism, reflected in immediate stock price declines following both announcements, has given way to analyst endorsement and subsequent equity recovery. I believe this suggests growing institutional confidence in AI-enabled acquisition strategies.
Innovation capital is flowing into companies that predate the viral “ChatGPT moment,” emphasizing that foundational work driven by the venture market has been in motion for years. For many of these AI-native companies founded nearly a decade ago, ChatGPT was the unlock for the elusive product market fit that all startups are in search of.
As AI becomes embedded in enterprise workflows and R&D processes, the ripple effects on employment and innovation patterns deserve careful consideration.
The startup-over-incumbent thesis
Two years ago, as I contemplated the anti-trust-driven deconstruction of the large cap tech incumbents protected by their walled gardens, I assumed private equity would play a role in this deconstruction. I did not anticipate the facilitators could be the well-funded venture backed startups.
Perplexity’s unverified $32 billion bid to acquire Chrome speaks to a new thesis: owning advanced AI-driven interfaces could shift how users discover and interact with information—flipping the script on tech incumbents.
Or, from a Perplexity point of view, perhaps this was an attempt at generating an intentional AI hallucination with no expectation of conversion, reinforcing the belief that “all news is good news.”
Adversaries turned allies
Meta and Alphabet’s $10 billion cloud computing partnership—an unexpected alliance between rivals—reflects the shared imperative to build AI infrastructure collaboratively, reinforcing that building the future may require unprecedented cooperation.
This reminds me of a tactic in Sun Tzu’s Art of War, whereby one uses strategy, diplomacy, and intelligence in the form of temporary and pragmatic measures to achieve mutual objectives. It will be interesting to see if other partnerships emerge among long-time adversaries as they contemplate how to fend off a very worthy list of new entrants.
While such collaboration may accelerate AI development, it also underscores the importance of ensuring these alliances serve broader innovation ecosystems, not just incumbent interests.
Connect with HarbourVest
Transitioning back to the future
As we reflect on these market dynamics from our investment committee discussions, much of what we are seeing is unprecedented. In the spirit of Dario Amodei’s provocation, today’s strategic acquirers are not discovering new continents—they are building them. Each $40 billion OpenAI round, each Meta partnership, each Siemens acquisition is less about claiming existing territory and more about constructing the very infrastructure upon which tomorrow’s economy will operate.
For investors, this presents both unparallelled opportunity and profound responsibility. The companies we back today, whether emerging AI natives or incumbents pivoting toward intelligence, are not merely building products. They are potentially architecting the foundational systems that will govern commerce, creativity, and human interaction for decades to come.
We, as a society, still have the opportunity to shape this vastly different future state and the opportunity for all of us to successfully evolve into it. We need to take that responsibility very seriously. The question for our industry is not simply how to generate returns from this transformation — it is how to ensure that the future we are funding serves not just shareholders, but society itself. Because these will be the rules of the game that we all need to live by.
Disclosure
HarbourVest Partners, LLC (“HarbourVest”) is a registered investment adviser under the Investment Advisers Act of 1940. This material is solely for informational purposes; the information should not be viewed as a current or past recommendation or an offer to sell or the solicitation to buy securities or adopt any investment strategy. In addition, the information contained in this document (i) may not be relied upon by any current or prospective investor and (ii) has not been prepared for marketing purposes. In all cases, interested parties should conduct their own investigation and analysis of information set forth herein and consult with their own advisors. HarbourVest has not acted in any investment advisory, brokerage or similar capacity by virtue of supplying this information. The opinions expressed herein represent the current, good faith views of the author(s) at the time of publication, are not definitive investment advice, and should not be relied upon as such. This material has been developed internally and/or obtained from sources believed to be reliable; however, HarbourVest does not guarantee the accuracy, adequacy or completeness of such information. The information is subject to change without notice and HarbourVest has no obligation to update you. There is no assurance that any events or projections will occur, and outcomes may be significantly different than the opinions shown here. This information, including any projections concerning financial market performance, is based on current market conditions, which will fluctuate and may be superseded by subsequent market events or for other reasons. The information contained herein must be kept strictly confidential and may not be reproduced or redistributed in any format without the express written approval of HarbourVest.
