Big Tech AI Agent Race Accelerates as SaaS Valuations Stabilize

The artificial intelligence agent market has shifted from speculation to execution. Microsoft officially launched Agent 365 for commercial customers on May 1, 2026, while Meta and Google are developing personal AI assistants to handle everyday tasks. The timing coincides with a stabilization in enterprise software valuations after the so-called "SaaS-pocalypse" wiped out approximately $2 trillion in market capitalization earlier this year.

According to DigitalToday reporter Chi-gyu Hwang, competition intensified after OpenClaw, an open-source AI agent tool, drew attention early in 2026. Meta is developing a personalized AI assistant called Hatch similar to OpenClaw, while Google is building a Gemini-based personal agent to support work, study, and daily life. The enterprise side is equally crowded: Microsoft's Agent 365 integrates and manages AI agents across Copilot, Teams, and Microsoft 365, including agents from external partners.

Microsoft's official documentation confirms Agent 365 is now generally available for commercial customers, designed to help organizations discover, govern, and secure AI agents operating across Microsoft, third-party SaaS, cloud, and local environments. The platform addresses what the company calls "agent sprawl" — the proliferation of autonomous systems that can invoke tools, access data, and interact with other agents. When an agent can execute tasks independently, any helpful workflow can turn into data oversharing or over-privileged actions in seconds (a problem that has plagued users for years, frankly).

The governance challenge extends beyond Microsoft's ecosystem. ServiceNow and Amazon Web Services expanded their partnership to include integrated AI agent governance and native development tools. ServiceNow is positioning itself as an AI agent governance platform spanning multiple platforms, strengthening links between an AI control tower and AgentCore to support multi-agent integration. Anthropic made Claude available within Microsoft Office, with Claude for Excel, PowerPoint, and Word reaching general availability while Outlook entered public beta.

The market context matters. In February 2026, approximately $2 trillion in market capitalization evaporated from the software sector as AI agents began replacing traditional SaaS tools. Atlassian dropped 35% after reporting its first-ever enterprise seat decline. Salesforce fell 28% despite revenue growth, as investors shifted focus from top-line numbers to declining net-new customer acquisition. The catalyst was AI agents performing tasks that previously required dedicated software interfaces, breaking the per-seat pricing model that powered SaaS valuations for over a decade.

However, enterprise software shares have recently recovered. The "SaaS big three" posted earnings surprises that appeared to brush aside SaaS-pocalypse concerns. Datadog and Twilio are emerging as software companies benefiting from AI. The narrative is shifting from "AI will replace SaaS" to "AI is transforming SaaS business models." Intercom is already exploring charging per customer resolution instead of per seat — a fundamental change in unit economics.

OpenAI replaced ChatGPT's default model with GPT-5.5 Instant, which reduces hallucinations in sensitive areas like law, healthcare, and finance while maintaining low latency. The company added three new voice models to its API with real-time conversation, translation, and transcription functions. OpenAI will expand its ChatGPT advertising pilot to five countries including South Korea within weeks. Perplexity opened its local AI agent, Personal Computer, to all Mac users. Adobe introduced an agentic AI solution for marketers called CX Enterprise Coworker. Deel launched an AI agent platform for practitioners called Akai.

The technical reality is that AI agents are becoming more capable and more integrated. Airbnb disclosed that 60 percent of the code written by engineers was generated by AI. Spotify launched a beta feature that lets users save and listen to personalized podcasts generated by an AI agent. LG CNS unveiled PhysicalWorks, a robotics transformation platform that integrates robot training and operations. SAP is acquiring AI research lab Prior Labs and data lakehouse platform Dremio, planning to invest more than 1 billion euros over the next four years.

What's actually surviving looks different. Companies building exclusively for one workflow or industry survive not because AI cannot reach them, but because the switching cost and domain depth make the economics unattractive for platform players. The vertical-first argument is solid: general capability is getting commoditized fast, but context, constraints, and the decisions around them are harder to replace. That's where the real moat is starting to form.

The per-seat to per-outcome shift is the real story. Seat-based pricing already felt stretched in 2022, but with agents, pricing by seats is actively punishing the vendor — you're capping revenue while the customer extracts 10x the value. Intercom moving to per-resolution is rational, but it's also an admission that the moat was never the UI — it was the workflow ownership. Vertical depth plus outcome-based pricing is the only durable combo left.

For enterprises, the message is that native controls from Microsoft or Google may help, but they are unlikely to cover the full agent landscape. Companies using multiple clouds, SaaS tools, developer platforms, and browser-based AI assistants will still need governance that extends beyond any single vendor's console. The harder problem is governing agents that operate beyond native platform boundaries — shadow AI agents can still emerge through developer tools, browser extensions, local assistants, SaaS copilots, and unsanctioned tool connections.

The practical takeaway: if your whole product is a better UI on top of Claude or ChatGPT, you don't have a defensible business. Every model update narrows your gap. The platform you built on is becoming your competition. What's actually surviving is software so vertical the AI giants won't bother — or tools that solve problems AI can't yet handle reliably. The rest is just waiting for the next API update to make your entire product obsolete (which, let's be honest, is already happening).

AI agents are not replacing SaaS. They're replacing undifferentiated SaaS. The era of building generic tools on top of someone else's AI is ending. The era of owning business logic, beyond interface, is beginning.

For more on Microsoft's Agent 365 capabilities, see the official Microsoft blog post. For market context on the SaaS-pocalypse, see Jon Radoff's analysis. For the original reporting on Big Tech's AI agent moves, see DigitalToday's coverage.

Artūras Malašauskas is an AI Systems Integrator with 20+ years of production-grade web engineering experience. He has designed, shipped, and scaled enterprise Python/PHP systems for logistics, SaaS, and public-sector clients. For the past year, he has focused exclusively on AI integrations: deploying open-source LLMs, building generative media pipelines (image, audio, video), and engineering multi-agent workflows for real production environments. His standard: reproducibility, security, cost-efficient inference—no vaporware. He documents and evaluates emerging AI tooling, separating verified capabilities from marketing noise. Technical editor at: muza-ai.eu, ai-verslas.lt, ai-naujinos.lt Connect on LinkedIn