Edited by Korea IT Times
Technology shares could remain turbulent in the months ahead as investors reassess how quickly artificial intelligence is reshaping the economics of software and services. The latest selloff reflects growing concern that more capable AI tools may weaken pricing power in legacy software models and disrupt existing revenue structures.
Recent declines in software and services equities have been sharp, with investors reacting to the idea that AI can perform a wider range of tasks that were previously bundled into premium-priced products. As a result, the market is shifting from broad enthusiasm about AI to a more selective debate over which business models can still defend margins.
This has produced a clear split within the tech sector. Companies tied to AI infrastructure—such as data-centre and computing supply chains—have generally held up better than firms reliant on recurring licences, long-term contracts, or workflow automation that can be replicated by newer AI-driven alternatives. In other words, markets are differentiating between businesses that sell the “picks and shovels” of AI and those whose value proposition may be easier to unbundle.
Broader market signals point to the same tension. Software-focused indexes across major regions have weakened as investors digest what generative AI can now do inside core enterprise functions. At the same time, risk-off positioning has surfaced in parts of the market as some capital has rotated toward traditional defensive assets amid equity volatility.
Nigel Green, CEO of deVere Group, argues that the repricing is less about hype fading and more about a practical test of pricing power in an AI-enabled economy: “The deeper issue markets are grappling with is economic: pricing power is now being adjudicated through AI’s ability to unbundle value, compress workflows, and deliver outputs with minimal human intervention.”
Two pressures are driving the repricing. First, AI can lower switching costs by reducing friction in replacing established tools and service providers, undermining the stickiness that historically supported long-duration software revenues. Second, investors are applying tighter standards to the gap between product ambition and monetisation, pushing companies to prove that AI spending can translate into durable earnings rather than temporary narrative support.
Volatility may persist as markets search for a new equilibrium—especially through earnings seasons, when guidance and margins will be tested against real AI adoption and competitive pressure. Firms that can demonstrate defensible economics—through proprietary infrastructure, unique data advantages, or genuinely scarce capabilities—may command a premium. Those that rely mainly on adding generic AI features to protect legacy pricing may face continuing scrutiny.
Macro and geopolitical conditions could intensify these moves. Interest-rate uncertainty, trade and tariff risks, and uneven growth expectations are making investors more selective, amplifying price swings when confidence in future earnings changes.
The adjustment may be uncomfortable, but it also reflects a more mature phase of market pricing: less faith in slogans, more focus on whether AI strengthens or weakens a company’s ability to sustain profits.
