Original reporting: Software companies fight back against fears that AI will kill them — Reuters, March 12, 2026

The SaaS industry just lived through its "oh shit" moment.

Last month, software stocks lost nearly $1 trillion in market value after Anthropic launched Claude Cowork — an AI agent that can automate the exact tasks that traditional software companies built their empires on. Customer relationship management. Business process workflows. Data organization. The whole stack.

I've written about Claude Cowork's enterprise implications before, but this week brought something more interesting than the initial panic: the counter-offensive.

Oracle, Salesforce, and Workday CEOs have all taken to their earnings calls to tell Wall Street — and their enterprise customers — why they're not toast. Some arguments are compelling. Some are desperate. And if you're an enterprise buyer trying to figure out which vendors to bet on for the next five years, you need to know the difference.

When $1 trillion evaporates, everyone suddenly has a defense strategy.

⚡ TL;DR: Software vendors are splitting into two camps: those with proprietary data moats (Oracle, Salesforce) and those whose data is standardized and replicable by AI (Workday). If you're locked into the wrong vendors, your migration costs just became strategic investments. Here's how to assess your portfolio before the market does it for you.

Oracle: "We're Building WITH AI, Not Getting Replaced BY It"

Oracle CFO Mike Sicilia delivered what might be the strongest defense I've heard so far. On Tuesday's earnings call, he said:

"You've all heard ... that new companies coding quickly using AI will spell the death of SaaS. I don't agree with that at all. I do think that AI tools and their coding capabilities would be a threat if we weren't adopting them, but we are, and very rapidly."

Let me translate that from executive-speak: "We're using AI to build faster than AI can replace us."

And the numbers back it up. Oracle predicted that the AI boom would power its revenue for several quarters, sending shares up 10% on Wednesday. The company sits on deep enterprise data across finance, supply chain, and human resources — the kind of data that AI can't easily replicate because it's locked inside Oracle's database and cloud infrastructure.

According to Rebecca Wettemann, CEO of tech research firm Valoir, Oracle's real advantage is flexibility: cheaper cloud systems plus a database that runs on any major cloud provider. "That flexibility gives customers choice — and that's a powerful position to be in as the AI ecosystem evolves," she told Reuters.

Translation for enterprise buyers: Oracle is betting that multi-cloud portability plus proprietary data integration beats pure AI automation. If you're already invested in Oracle's stack, they're in a stronger position than most to defend that investment.

But Sicilia also took a shot at Salesforce, claiming Oracle was "using AI to actually build new products and automate full business processes, not just add AI features on top of existing tools."

That's a direct challenge. So what's Salesforce's counter?

Salesforce: "Our Data Moat Is Deeper Than Your AI Agent"

Salesforce CEO Marc Benioff didn't just defend his company last month — he brought customers on stage to do it for him.

The pitch: Salesforce isn't a CRM company anymore. It's an enterprise platform for building, deploying, and governing AI agents, powered by what Benioff called "mountains of proprietary customer and sales-process data."

Let me give you the specific number that matters: Salesforce's real-time data platform manages more than 50 trillion records. That's not marketing fluff. That's decades of unique, relationship-driven, context-rich data that no AI startup can replicate by scraping the web or training on synthetic datasets.

When your moat is measured in trillions of proprietary records, AI disruption looks different.

Here's why that matters from an enterprise buyer perspective: not all data is equal.

Nearly a dozen tech analysts and investors surveyed by Reuters said the same thing: companies with years of exclusive financial, legal, design, or technical data have the best defense against AI disruption.

James St. Aubin, chief investment officer at Ocean Park Asset Management, put it bluntly: "Proprietary data is the deepest moat by far."

Salesforce's data isn't just large — it's unique. CRM data captures relationships, negotiation history, deal structures, and buying patterns that are specific to each business. That's contextual intelligence, not commodity information.

Compare that to HR and payroll data, which follows standardized formats across industries. An AI agent trained on enough payroll systems can replicate most of what Workday does. But an AI agent can't replicate your sales team's relationship history with a Fortune 500 account — not unless it has access to your Salesforce instance.

The enterprise risk question: Is your data a commodity or a competitive advantage? If it's standardized, your vendor is vulnerable. If it's proprietary and contextual, you have leverage.

Workday: "AI Can't Replace Our Business Logic... Yet"

And then there's Workday.

The company brought back co-founder Aneel Bhusri as CEO last month to lead the company "in the rapidly evolving AI era." Shares have dropped more than a third this year, hitting a five-year low after sluggish sales forecasts.

Bhusri's defense on the earnings call was... let's call it philosophical:

"AI, for all of its incredible capabilities, is probabilistic by nature. It reasons, predicts and recommends based on patterns and likelihoods. Maybe it will eventually become a state machine — a system that follows the same steps and gets the same result, every time — but it is not there today."

Translation: "AI isn't deterministic enough to replace us... yet."

Here's the problem with that argument: Workday's core products run on HR and payroll data, which analysts repeatedly described as following "uniform, industry-standard formats." That means an AI company can more easily learn from — or replicate — tools built on that kind of data.

I've talked to enterprise procurement leaders who are already testing AI-native HR tools that can generate payroll reports, compliance filings, and benefits summaries using generic payroll schemas plus natural language prompts. The switching costs are still high, but they're dropping fast.

The brutal reality: Workday's moat isn't proprietary data. It's decades of embedded business processes and the pain of migration. That's a cost barrier, not a competitive advantage. And AI is actively eroding cost barriers across enterprise software.

When your defense is "migration is hard," you're not defending — you're delaying the inevitable.

The Data Hierarchy: Which Vendors Have Real Moats

If you're responsible for vendor strategy, here's the framework I'd use to assess your portfolio:

Tier 1: Proprietary, Contextual Data (Strong Defense)

• Salesforce — relationship data, deal history, customer interaction patterns
• Oracle — integrated financial/supply chain/operational data across enterprise systems
• ServiceNow — incident history, workflow patterns, organizational knowledge graphs

Why they're safer: The data isn't replicable without access to your specific business operations. AI can build a CRM, but it can't build your CRM without your Salesforce data.

Tier 2: Standardized but Deeply Embedded (Moderate Risk)

• Workday — HR/payroll follows industry standards, but migration is expensive
• SAP — ERP data is standardized, but process integration is complex
• Adobe Experience Cloud — marketing data patterns are replicable, but toolchain lock-in is real

Why they're riskier: AI can replicate the functionality, but the switching costs buy time. These vendors need to evolve fast or risk commoditization within 3-5 years.

Tier 3: Commodity Workflows (High Risk)

• Generic CMS platforms
• Basic HRIS tools without deep process integration
• Standalone analytics dashboards (easily replaced by AI-native BI)

Why they're vulnerable: If an AI agent can do the job with a prompt, your vendor is dead. They just don't know it yet.

Jensen Huang's Reality Check

NVIDIA CEO Jensen Huang — who arguably understands AI economics better than anyone — called fears that AI would replace software "illogical" last month.

His argument: AI doesn't replace software. It transforms how software is built, deployed, and consumed. The companies that adopt AI tooling to accelerate their own development cycles will win. The ones clinging to legacy business models will lose.

I'd add one nuance: AI replaces commodity software. It augments differentiated software.

If your vendor's product could be described in a three-sentence prompt and replicated by Claude or GPT-6, that's a dead vendor walking. If it takes years of proprietary data integration and domain expertise to deliver what they deliver, they've got a shot.

What Enterprise Buyers Should Do Right Now

I'm not saying to rip out Workday tomorrow or go all-in on Oracle. But I am saying this: the next 18-24 months are your renegotiation window.

Here's the playbook I'd recommend for any enterprise IT leader:

1. Audit Your Vendor Portfolio by Data Moat Strength

Build a simple matrix:

• Column 1: Vendor name
• Column 2: Data moat (proprietary/standardized/commodity)
• Column 3: Migration cost (high/medium/low)
• Column 4: AI replacement risk (low/medium/high)

If a vendor is standardized data + low migration cost + high AI risk, you need a contingency plan by Q3 2026.

2. Negotiate AI-Proof Contract Terms

Every renewal should include:

• Feature parity clauses — if the vendor adds AI capabilities, you get them at no upcharge
• Data portability guarantees — 90-day export windows with API access, not just CSV dumps
• Price protection — if AI reduces the vendor's operating costs, you get a corresponding discount

The vendors who push back on these terms are telling you they don't believe their own AI defenses.

3. Test AI-Native Alternatives in Parallel

Don't wait for your vendor to get disrupted. Run pilot projects with AI-native competitors now while you still have leverage.

For Workday customers: test AI-powered HR tools like agentic banking systems are doing in pilot phases.

For Salesforce customers: evaluate whether AI agents can handle routine CRM workflows before Salesforce forces you into Agentforce pricing.

For Oracle customers: honestly, you're probably fine for the next 3-5 years. But track infrastructure dependencies and vendor concentration risk closely.

4. Build Multi-Vendor Resilience

The OpenAI-Oracle Stargate collapse proved that even $500 billion commitments can fracture. Single-vendor dependency is a strategic liability in the AI era.

Run critical workloads across at least two platforms. Yes, it costs more. No, you won't regret it when one vendor gets disrupted or acquired.

The Bottom Line: Bet on Data Moats, Not Brand Names

The $1 trillion SaaS rout wasn't irrational panic. It was the market pricing in a real risk: AI is coming for commodity software.

Oracle and Salesforce have credible defenses built on proprietary data and integrated platforms. Workday is hoping business process complexity buys them time. But time runs out.

If you're an enterprise buyer, your job isn't to pick the winner. It's to not be locked into the loser when the market figures it out.

Audit your vendors. Negotiate flexibility. Test alternatives. And remember: in a disruption cycle, the companies that move first pay the least.

The trillion-dollar question isn't whether AI will kill software companies. It's which ones — and whether you're ready when it happens.

---

Want to calculate your own AI ROI? Try our AI ROI Calculator — takes 60 seconds and shows projected savings, payback period, and 3-year ROI.

Continue Reading

Vendor Risk & AI Strategy:

• The Government Just Cut Off Anthropic Overnight — When vendor risk becomes geopolitical risk
• OpenAI and Oracle Just Blew Up Their Biggest AI Data Center Deal — What $500B infrastructure fractures mean for buyers
• AI Agents in the Enterprise: 2026 Adoption Reality Check — Where the automation actually works today

---

Know someone who'd find this useful?

Forward this article to a colleague who's navigating vendor strategy in the AI era. They can subscribe at beri.net/#newsletter — it's free, twice a week, and I read every reply.

Connect with me:

• LinkedIn
• Twitter/X
• Facebook

---

Continue Reading