The massive capital expenditure in AI is largely confined to the "superintelligence quest" camp, which bets on godlike AI transforming the economy. Companies focused on applying current AI to create immediate economic value are not necessarily in a bubble.
In a new technological wave like AI, a high project failure rate is desirable. It indicates that a company is aggressively experimenting and pushing boundaries to discover what provides real value, rather than being too conservative.
LLMs are becoming commoditized. Like gas from different stations, models can be swapped based on price or marginal performance. This means competitive advantage doesn't come from the model itself, but how you use it with proprietary data.
The massive investment in AI seems disproportionate to the software market's size. However, its true potential is in automating and augmenting the services industry, which is 25 times larger than software, thus justifying the spend.
A significant hurdle for AI, especially in replacing tasks like RPA, is that models are trained and then "frozen." They don't continuously learn from new interactions post-deployment. This makes them less adaptable than a true learning system.
The tedious manual process of data entry into systems like Salesforce is ripe for disruption. AI agents that analyze meeting recordings (e.g., from Zoom) to automatically extract action items and update records are already emerging as a key use case.
Since LLMs are commodities, sustainable competitive advantage in AI comes from leveraging proprietary data and unique business processes that competitors cannot replicate. Companies must focus on building AI that understands their specific "secret sauce."
Today's AI models have surpassed the definition of Artificial General Intelligence (AGI) that was commonly accepted by AI researchers just over a decade ago. The debate continues because the goalposts for what constitutes "true" AGI have been moved.
The AI landscape has three groups: 1) Frontier labs on a "superintelligence quest," absorbing most capital. 2) Fundamental researchers who think the current approach is flawed. 3) Pragmatists building value with today's "good enough" AI.
In the current, rapidly evolving AI market, the long-term winners are not yet clear. CIOs should de-risk their budgets by experimenting with more vendors, using shorter-term contracts, and prioritizing products that can be tested and prove value quickly.