GeekWire reached out to a select group of Seattle venture capitalists to gather their perspectives on the AI market and the potential for a bubble. While they acknowledge areas of overheating, particularly in early-stage startups where valuations may exceed real traction, most do not foresee a catastrophic bubble. The general consensus is that AI is already proving its value, despite some excesses in the market. Each investor brings a unique viewpoint, with some highlighting concerns about data center buildouts and narrative-driven startups with inflated valuations, while others emphasize the immediate opportunities for companies reevaluating their software spending.
The overarching advice from these investors to startup founders is to stay grounded, focus on solving tangible customer problems, establish sustainable revenue streams, and prepare for potential market adjustments.
Sabrina Albert (Wu), partner at Madrona
“There’s evident excitement in certain segments of the AI market, particularly in the realm of early-stage private valuations where companies are often priced beyond their actual performance, aligning with a classic ‘bubble’ scenario. While publicly traded AI firms show strong valuations supported by robust earnings and growth, the private market, especially at the seed and Series A stages, witnesses excessive exuberance with investors eager to gain early exposure to AI. As a result, capital is flowing towards startups with limited traction and valuations that anticipate outcomes requiring years of execution to validate.
Startup strategies should prioritize establishing strong foundational business principles from the outset. Focus on generating consistent revenue through long-term contracts, address genuine customer needs, and differentiate by deeply integrating into customers’ technology stacks to create sustainable product and company growth trajectories. Long-term success hinges on delivering measurable value and fostering defensible growth over time.”
Cameron Borumand, general partner at Fuse
“Various factors are in play within the AI landscape. The emergence of AI as a truly transformative technology will reshape industries significantly in the long term. However, historical patterns indicate that new technologies tend to be overhyped initially and underestimated over time. The profound impacts of AI may only fully materialize in 10 to 20 years.
In the near future, we may witness a pullback in public markets as investors come to terms with the reality that achieving ‘enterprise readiness’ for AI will require time. This adjustment does not imply a catastrophic event but rather a correction from the exceptional 21% year-over-year growth in the Nasdaq, possibly reverting closer to the historical 10% average. Despite some overhyped companies in the late-stage private markets, the tangible impact of AI is already evident, as seen in companies like Anthropic scaling their revenue significantly.
For startups, the current environment presents an opportune moment to innovate. M&A activities are resuming, customers are allocating budgets, and talent is eager to engage in compelling projects. Amidst the noise, it is crucial to focus on solving core customer challenges deeply. While infrastructure has dominated growth thus far, the upcoming years will spotlight the next wave of AI-powered applications.”
Chris DeVore, founding managing partner at Founders’ Co-op
“Indeed, a substantial portion of global capital deployed in AI, particularly in data center expansions, is likely misallocated. Specifically within startups, except for a few frontrunners (OpenAI, Anthropic, Cursor), the concern lies not in excessive capitalization but in the valuation of companies relative to their actual cash flows and margin potential.
Unlike previous bubbles like crypto and metaverse, the current scenario retains genuine value within Language Model Models (LLMs) despite their current developmental stage. LLMs are potent tools even in their current state and are poised to remain integral to various software development and knowledge work tasks in the foreseeable future.
The challenge for founders and investors during such periods is to make decisions that will prove wise a decade from now, not just in the present moment. Is there a way to leverage LLMs to create enduring business value in sectors less likely to suffer from overcapitalization or fierce competition? Alternatively, attempting to identify winners amidst the capital frenzy and paying whatever price the market demands for those assets is a risky strategy, even for seasoned players.
The recipe for success in such times mirrors that of any other era: identify a customer segment you understand better than anyone else, deeply engage with these customers to identify problems that LLMs can uniquely solve, rapidly iterate to showcase value, and sustain this pace of innovation and learning over time. While it may sound straightforward, very few founding teams can execute this effectively, underscoring the challenges and excitement inherent in building startups.”
Sheila Gulati, managing director at Tola Capital
“Overall, I do not perceive the current state of the AI landscape as indicative of a bubble. Similar concerns arose when we introduced the Azure platform approximately fifteen years ago, with initial apprehensions about plunging into a zero-margin business model.
The extensive AI infrastructure expansions underway will redefine operational software layers essential for real-world performance, encompassing compute orchestration, data pipelines, memory systems, and large-scale inference efficiency. Value is shifting towards packaging and deploying intelligence across enterprise workflows.
Enterprise software startups should position themselves within the expanding Total Addressable Market (TAM) by providing comprehensive end-to-end solutions and pioneering new operational paradigms where human-AI collaboration thrives. Successful startups will straddle the burgeoning IT TAM and tap into a segment of the labor market’s economics.
The CIO budget flexibility witnessed today is unprecedented. Established application stacks are now ripe for disruption by AI-centric newcomers. The market presents immense opportunities, urging companies to aim for becoming new industry giants rather than minor players focused on isolated features.”
Andy Liu, co-founding partner at Unlock Venture Partners
“Undoubtedly, we are amidst an AI bubble, albeit not in the conventional sense. Capital and valuations are surpassing actual performance levels, particularly for companies lacking substantial customer traction, unique differentiation, or viable paths to profitability. The gap widens between narrative-driven AI enterprises, where ‘AI’ serves primarily as a marketing tool, and value-oriented firms leveraging AI to deliver measurable, consistent customer value.
The bubble’s intensity is most pronounced in the early and growth stages, where AI narratives briefly mask the absence of significant traction, enabling startups to secure funding at inflated valuations. While some robust companies will emerge from this cycle, others may face substantial downturns, restructurings, or closures as they struggle to meet lofty expectations.
Looking towards 2026, my advice to founders is straightforward:
- Focus on building tangible businesses, not just impressive presentations. Develop products capable of generating actual revenue before seeking investments.
- Emphasize efficiency, customer Return on Investment (ROI), and sustainable unit economics.
- Leverage AI to create genuine leverage, rather than using it as a justification for excessive spending.
The upcoming year presents an exceptional opportunity for building innovative products. The cost of experimentation and product development has decreased significantly, enabling founders without traditional educational credentials to create viable products and revenue streams. The next wave of enduring AI companies will likely emerge from small teams prioritizing efficient execution over hype. We eagerly anticipate witnessing more teams develop remarkable products in the coming year.”
Annie Luchsinger, partner at Breakers
“In my view, what we are witnessing is not merely an AI bubble but rather a classic venture cycle unfolding amidst a truly transformative platform shift. Venture capital has historically adapted to new paradigms alongside significant technological advancements (cloud, mobile, social), with AI marking the fastest-paced shift to date.
The distinguishing factors this time are the rapidity, scale, and availability of capital. AI adoption is occurring at an accelerated pace and on a grander scale than previous platform transitions, coinciding with unprecedented levels of private-market capital. As these dynamics converge, pricing, timelines, and investor behavior undergo evolution.
The trend of capital outpacing real-world performance is not novel. While some adjustments and shakeouts may occur, genuine value creation persists. Companies armed with authentic technology, robust distribution channels, and loyal customer bases will endure the market shifts.”
