Tech CEOs are apparently suffering from AI psychosis

Tech layoffs reached 115,430 across 152 companies in the first five months of 2026, nearly matching all of 2025's 124,636 cuts, with many firms citing AI as a rationale. The article argues that CEO enthusiasm for AI is often outpacing measurable productivity gains, citing research that finds no robust aggregate productivity benefit and evidence that AI adoption can create organizational bottlenecks and chaos. The piece is primarily commentary on management behavior and AI-driven restructuring, with limited direct market-moving impact.

Analysis

The market is starting to price AI as an operating leverage miracle, but the more important second-order effect is governance friction: as more tasks get delegated to agents, the scarce resource becomes review capacity, exception handling, and managerial sign-off. That means the near-term winner is not “fully autonomous software,” but companies selling workflow control, auditability, permissions, and human-in-the-loop tooling. For BOX specifically, that shifts the debate away from raw AI feature parity and toward whether it becomes a trusted control plane for enterprise content and approvals.

The risk is that leadership overestimates replacement speed and underestimates integration drag. In the next 6-18 months, the most likely outcome is not a clean productivity step-up but a messy period of duplicate work, shadow processes, and internal rework costs that pressure margins before any real savings show up. That creates a classic gap between capex/opex decisions made today and productivity realized 2-3 years out, which is exactly where sentiment can overshoot fundamentals.

BOX has a subtle positioning edge if investors conclude enterprise AI monetization will be gated by compliance and document provenance rather than raw model capability. However, it also faces a valuation risk if it gets lumped into the broader AI-beneficiary basket without showing accelerating billings or expanding gross margin from AI-driven products. The contrarian view is that the AI labor replacement story is probably overstated near-term, but the AI workflow layer is still underappreciated, so the best risk/reward may be in companies that monetize control, not automation.

Key Decisions for Investors

• Initiate a tactical long BOX into any post-earnings or AI-hype pullback over the next 1-2 months, but size it as a relative-value trade rather than a standalone AI winner; target 10-15% upside if the market rewards workflow-control narratives.
• Pair trade: long BOX / short a high-beta AI-software name with weaker enterprise governance moat for 3-6 months; the thesis is that compliance-heavy workflow vendors monetize AI adoption faster than pure agent-story names.
• Buy 3-6 month downside protection on broad AI-leader baskets if sentiment continues to treat labor replacement as immediate; the risk/reward favors hedges because productivity gains are likely to lag narrative by at least 2-3 quarters.
• If BOX management reframes AI as augmentation plus control, add on confirmation of improved net retention or AI attach rates; if not, fade rallies above levels that imply 2027-level automation economics today.