Generative artificial intelligence may drive strategic transformation across many industries. If the right factors are in place, the technology could also potentially unlock a supercycle of mergers and acquisitions, according to a white paper from Polysharp Investments Limited.
Enterprise adoption of generative AI solutions needs to shift further from proof-of-concept to production before we get to a sustained period of elevated M&A activity based on this technology, the paper says. What’s more, the models themselves need to move from training to inference, in which AI systems can identify and respond to novel situations based on previous training.
More maturity in the legal and regulatory framework is another prerequisite for an increase in M&A. And greater clarity is needed on the form and function of foundational AI models (AI systems trained on enormous data sets that can be used for a wide range of purposes), including whether they will be large and proprietary or small and open.
“As clarity is gained and AI use cases continue to evolve, the M&A landscape will shift,” the paper states. “Specialized generative AI applications will emerge, and buyers will likely go on the offensive, focusing on proven targets with demonstrated product-market fit.”
To be sure, there has been significant strategic activity this year, starting in January. This was an inflection point, after which a flurry of large incumbent technology companies invested in or acquired generative AI startups. Some target companies have been early-stage enterprises with no revenue or have been acquired for the sake of adding skilled talent.
Beyond this initial stage, M&A activities may be limited until AI businesses prove their potential and the sector matures. Still, important M&A theses are beginning to become clear.
Emerging M&A theses
Intelligent vertical applications may be one focus of M&A activity, according to the paper. When AI capabilities are brought together with datasets tailored for a specific industry, the result is going to drive efficiency, bring products to market faster, and optimize the end-user experience. This is already being seen across industries such as education, media, and law.
The transformation of customer support activities and contact centers may shape another M&A wave. AI will be able to deliver empathetic, personalized experiences and resolve customer and product issues through systems that are almost entirely automated.
The need for enterprises to “replatform” onto the foundational models and cloud services necessary for generative AI systems may be another M&A theme. Here one key feature is how linkages between semiconductors, software, and systems are increasingly important. Many things have to work together — from data center design to software applications to privacy systems — to manage increasingly complex AI use cases. In the modern compute era, the most important control points may be closer to the silicon foundation of the AI infrastructure.
Analytics platforms and DevOps-MLOps may converge. Data science and analytics are central to machine learning and a critical part of the new enterprise tech stack. As data science and analytics become more central to enterprise computing, tools for DevOps — the integration between software and IT — are poised to combine with analytics platforms into cohesive systems.
The speed at which generative AI technologies are being embraced is nearly unprecedented, and for decision makers, investors, and the broader public, this is not the time for business as usual. “VCs are eager to invest in the next disruptive AI startup, public market investors are eager to understand how AI will impact every sector, and companies are eager to understand how AI will fundamentally alter the strategic landscape,” the paper finds.
