• 3 min read
Workday says the SaaS apocalypse is overblown
Analysts say agentic AI will disrupt enterprise software, but vendors like Workday, Freshworks, and Snowflake argue trust and core systems still matter.
Image: ZDNET
Agentic AI is forcing a rethink of enterprise software, but analysts and vendors say talk of a full-blown “SaaS apocalypse” goes too far.
According to Gartner, up to $234 billion in application spending could be exposed to agentic arbitrage between now and 2030, as AI agents complete work across multiple systems and reduce reliance on traditional software interfaces. By the end of the decade, Gartner says, this model could account for about 20% of enterprise application SaaS spending.
Those projections have fueled fears that AI agents will replace conventional software tools. Some estimates suggest markets have already lost roughly $300 billion in SaaS valuations over the past 18 months as investors price in that risk.
Still, IDC research director Shannon Kalvar told ZDNET the shift looks more like disintermediation than extinction.
Recommended reading
a16z backs Runta with $20M to rein in AI agents
“The SaaS apocalypse is overrated, but your use of the word disintermediation is correct.”
Kalvar said users may increasingly ask tools such as Claude, ChatGPT, or Perplexity to assemble temporary, task-specific software experiences on the fly, pulling in capabilities from multiple systems. That makes the traditional application interface less central. For software vendors, he said, the key question is whether their underlying capabilities will still be distinctive in five years.
How vendors plan to stay relevant
At Workday, that answer centers on trust, security, and cross-business workflows. Clare Hickie, CTO for EMEA at Workday, told ZDNET that many digital leaders remain cautious about AI adoption because of trust concerns.
“What’s crucial to recognize is that privacy by design and security frameworks are built into absolutely everything that we do.”
Speaking at Workday’s innovation media event at its EMEA HQ in Dublin, Hickie said the company is building what executives described as the “front door to work.” Through its next-generation service Sana, employees would log into Workday, ask natural-language questions about issues such as payroll variations, and get personalized answers from enterprise data sources.
Freshworks is making a similar case. At its Refresh 2026 event in New York City, CTO Murali Swaminathan said AI may become the engagement layer, but the operational data beneath it will still live in systems of record like Freshworks.
He argued that firms such as OpenAI and Anthropic are unlikely to build every category of enterprise software themselves, especially areas like service management support that depend on workflows, SLAs, and business rules.
Snowflake points to specialization
Snowflake co-founder and president Benoît Dageville made a related argument at the company’s Summit 2026 event in San Francisco. His position: AI companies may move into adjacent areas, but deep product specialization still matters.
“We have the data, and there is no AI without data.”
Dageville compared the moment to earlier fears that cloud giants such as Amazon would overrun specialist software vendors. He pointed to Amazon Redshift and Snowflake’s eventual relationship with Amazon Web Services, where Snowflake runs on AWS infrastructure and AWS customers can use Snowflake services.
His conclusion was blunt: Anthropic or OpenAI could try to become Snowflake competitors, but they do not automatically have an advantage in building data platforms. In his view, they may not even want to.
Enterprise Editor
Marcus follows the money. He covers enterprise software, cloud architecture, and the tectonic shifts in Big Tech strategy. He translates dense earnings calls and complex M&A activity into actionable insights about where the industry is actually heading. If a tech giant makes a silent pivot, Marcus is usually the first to notice.
via ZDNET
