Is Software Dead?

What It Covers

Markets experience significant sell-offs in software stocks as AI coding capabilities trigger fears about SaaS business model viability. Salesforce down 21%, Snowflake 23%, HubSpot 36% year-to-date. Debate centers on whether AI agents will replace traditional software or simply transform pricing models and competitive dynamics in enterprise technology.

Key Questions Answered

• •Market repricing mechanics: SaaS stocks face first-ever sector-wide disruption pricing, distinct from previous AI hype cycles. Apollo Global Management reduced software exposure from 20% to 10% in private credit funds while actively shorting positions. The sell-off represents fundamental questions about software longevity rather than temporary market jitters, with investors questioning whether products remain relevant long enough for traditional financial engineering to work.
• •Per-seat pricing crisis: Traditional SaaS revenue model faces existential threat as AI enables 10 people to accomplish work previously requiring 100 seats. High-growth, low-profitability strategies no longer attract investment. Companies must demonstrate clear profitability paths by 2026 or face investor exodus. Inference costs squeeze traditional high margins while AI capabilities fundamentally challenge the per-user licensing model that underpinned software industry growth for decades.
• •Enterprise reality gap: Large organizations operate on decades of layered systems including ERP, mainframes, custom services, and compliance controls requiring 12-month change plans. Stock prices move on expectations while enterprise architecture moves on risk tolerance, creating timing mismatches. 50,000-person industrial companies unlikely to vibe-code replacements for mission-critical systems like Workday, suggesting disruption timeline differs dramatically between nimble startups and established enterprises with complex technical debt.
• •Competitive moat differentiation: AI strengthens companies with distribution, proprietary data, workflow integration, enterprise lock-in, network effects, and compliance trust while destroying companies whose only moat was software itself. Strong software vendors can absorb arbitrary AI investment to improve product quality and compete on user experience. Weak vendors face commoditization as AI agents select optimal tools dynamically rather than maintaining long-term vendor relationships based on switching costs.
• •Agent-first transformation path: Public SaaS companies can survive through three-step transformation: dramatically cut stock-based compensation, aggressively deploy AI agents internally for efficiency gains, and transition products from traditional SaaS to agent-based revenue models. Companies maintaining customer relationships while adding AI capabilities position better than pure-play software vendors. The shift suggests 10x software usage in a decade but with fundamentally restructured pricing, procurement processes, and competitive landscapes.