AI Brand Strategy Drives Investment Through Hype and Rebranding

The AI Brand: A History of Hype, Winter, and Rebranding

In this conversation, historian Thomas Haigh unpacks the enduring, often misleading, history of "Artificial Intelligence" as a brand. The core thesis is that AI, from its inception, has functioned more as a marketing term than a stable scientific discipline, cycling through periods of intense hype and subsequent disillusionment. The non-obvious implication is that the current AI boom, driven by large language models, is not a radical departure but a re-appropriation of an old brand, leveraging science fiction narratives to secure massive investment. Those who understand this historical pattern gain an advantage by recognizing the cyclical nature of AI promises and the underlying technological and commercial forces at play, allowing them to cut through the hype and identify durable innovations. This analysis is crucial for investors, technologists, and anyone seeking to understand the true trajectory of AI beyond its current marketing frenzy.

The "AI Winter" Myth: A Tale of Elite Labs and Shifting Narratives

The conventional wisdom surrounding AI history paints a clear picture: periods of fervent progress punctuated by devastating "AI winters" where funding dried up and research stalled. However, historian Thomas Haigh challenges this narrative, arguing that the widely accepted story of multiple AI winters, particularly in the 1970s, is a distortion. Haigh contends that this perception stems from the limited perspective of a few elite labs at institutions like MIT and Stanford, whose funding from agencies like DARPA did indeed decrease.

"The received wisdom has become that AI did well in the 60s, then it did badly in the mid and late 70s. It did well again in the 80s around expert systems and the first time that AI is really having startups and venture capital and industrial enthusiasm in the 80s, and that's absolutely true. And then it does badly in the 90s and early 2000s before it revives around neural networks. And the part of that I disagree with is I think AI was doing pretty well in the 70s outside some very specific factors around DARPA..."

Haigh's research, by examining broader metrics like membership in AI associations and international conference participation, suggests that the 1970s were a period of growth and international expansion for AI, not decline. The "winter" was more of a localized frost affecting a few dominant labs, while the field itself was blossoming elsewhere. This suggests that the narrative of AI winters is not solely a reflection of technological stagnation but also a product of how influential figures within the field chose to frame their experiences, often shaping the historical record to their own perspectives. The consequence of this skewed narrative is a misunderstanding of AI's true developmental arc, potentially leading to an overestimation of current breakthroughs or an underestimation of past resilience.

The Brand That Wouldn't Die: AI as Marketing

From its very inception, the term "Artificial Intelligence" was a marketing tool. Coined by John McCarthy in 1955 for a proposal to secure funding for a summer school at Dartmouth College, the brand was designed to attract attention and resources. Haigh argues that this brand-like quality has been a defining characteristic of AI throughout its history, influencing what technologies are included under its umbrella and how they are perceived.

The history of neural networks provides a prime example. Initially a focus in the early days of AI, they were later pushed out of the "AI brand" and rebranded as pattern recognition, then machine learning, and finally deep learning. This rebranding allowed the core ideas to continue developing outside the shadow of AI's overpromises and subsequent "winters." The current resurgence of AI, driven by deep learning and large language models, represents a re-appropriation of the AI brand by the machine learning community. This move, Haigh suggests, is a deliberate strategy to tap into the science fiction-infused cultural narrative surrounding AI, which promises rapid, transformative, and even superhuman intelligence.

"So a question is, why did the machine learning community seize the AI brand for itself after having kept a distance from it for many years? And I think the argument for that is fundamentally around the science fiction narrative associations of AI."

This rebranding allows companies to attract investment and public attention by evoking the grand promises of AI, even if the underlying technologies are more specialized than the brand suggests. The consequence of this branding strategy is a continuous cycle of hype, where the term "AI" becomes detached from specific technological capabilities and instead represents a broad, aspirational, and often exaggerated vision of the future. This makes it difficult for the public and even many within the industry to discern genuine progress from marketing spin.

Expert Systems: The Pragmatic Pivot and the Illusion of a Second Winter

The 1980s saw a boom in AI, largely fueled by the rise of expert systems. This was a strategic pivot, moving away from the ambitious goal of achieving general human-level intelligence towards creating systems that could encapsulate specialized domain knowledge. Ed Feigenbaum, a key figure in this movement, championed expert systems as a way to make computers economically valuable by encoding expertise. The logic was that simulating a high-level expert in a narrow field was more achievable than replicating general intelligence.

This focus on practical applications and economic viability meant that even as some labs struggled with funding, the expert systems boom provided a lifeline and a sense of progress. Haigh points out that figures like Feigenbaum did not perceive the 1970s as a period of retrenchment because their work on expert systems was flourishing. This highlights how the "AI winter" narrative is often tied to specific research agendas rather than a universal slowdown. The success of expert systems, while not achieving the grand visions of early AI proponents, demonstrated that focused, knowledge-based systems could deliver tangible value. This pragmatic approach, though less glamorous than the pursuit of AGI, laid crucial groundwork for future AI development by demonstrating the power of encoded knowledge and inference engines. The consequence of this successful pivot was that the field continued to evolve, albeit under a different banner, avoiding a complete collapse and setting the stage for the next wave of AI innovation.

The Long Game: Delayed Payoffs and Durable Advantage

The history of AI is replete with instances where immediate solutions failed to deliver long-term value, while approaches requiring patience and upfront investment eventually yielded significant advantages. The current AI landscape, dominated by massive compute power and data, is a testament to this. Haigh's analysis implicitly suggests that the technologies that eventually succeeded, like neural networks, often required a sustained period of development and a willingness to explore approaches initially deemed less promising.

The current focus on large language models, while impressive, is built on decades of research in areas like neural networks and parallel computing. The Strategic Computing Initiative in the 1980s, for example, funded not only expert systems but also parallel computing and autonomous vehicles, technologies that have proven essential for modern AI. The consequence of understanding this long game is a strategic advantage. Instead of chasing the latest hyped AI application, one can focus on foundational technologies and approaches that have demonstrated resilience and potential for sustained growth. This requires a willingness to invest in areas with delayed payoffs, a trait often lacking in the fast-paced world of technology and venture capital. The historical pattern shows that true breakthroughs often emerge from persistent, foundational work, not just from rebranding existing technologies.

Key Action Items

• Reframe "AI" as a Brand: Recognize that "AI" is a marketing term that has evolved over time, encompassing diverse technologies. Focus on the specific capabilities rather than the overarching brand.
• Investigate Foundational Technologies: Prioritize understanding and investing in core technologies like machine learning, neural networks, and parallel computing, which have demonstrated long-term viability, rather than chasing ephemeral AI trends.
• Critically Evaluate Hype Cycles: Be skeptical of grand, science-fiction-like promises surrounding AI. Understand that historical AI narratives have often been driven by marketing and funding needs, not just technological reality.
• Focus on Specific Applications: Instead of pursuing general intelligence, identify and develop AI solutions for specific, well-defined problems where clear value can be demonstrated.
• Embrace Long-Term Investment: Recognize that significant technological advancements often require sustained effort and delayed gratification. Be prepared to invest in research and development with long-term payoff horizons.
• Distinguish Marketing from Substance: Develop the ability to differentiate between technologies that are genuinely novel and those that are simply rebranded or repackaged existing capabilities under the "AI" umbrella.
• Learn from Historical Cycles: Study the patterns of AI hype, investment, and disillusionment to anticipate future trends and avoid repeating past mistakes. This pays off in 12-18 months by providing a more grounded perspective on AI's potential.