The Morning Download: Adobe, Salesforce Unveil AI Plans Amid Threat of Disruption

Good morning. The threat of new AI tools from companies like Anthropic and OpenAI has spooked investors for months, wiping out $300 billion off software and data stocks.

The WSJ Leadership Institute's Belle Lin looks at Adobe. The company was an early beneficiary of AI hype with its homegrown Firefly AI models, but the sector's stock decline has proved inescapable—stock is down roughly 30% this year.

Adobe this week is making announcements tied to its latest efforts to stay ahead of AI-driven software disruption. Among them is CX Enterprise, an AI agent-based platform that can be used to help businesses boost areas like customer engagement, sales and loyalty.

“CX Enterprise is 100% a massive step toward saying, ‘Hey, this is how you can use our technology, which we see as incredibly vital in this AI era,’” Amit Ahuja, Adobe’s senior vice president of product for customer experience orchestration, tells Belle. “None of us have our heads in the sand as to not see what the [investor] questions are.”

Why Adobe and peers may weather the storm

Many corporate technology leaders have said they’re not immediately ripping out their software in favor of new AI tools. Plus, analysts and industry experts say software vendors hold critical business data on behalf of their customers, and that function isn’t easily replaceable.

The same pressures are mounting at Salesforce, another poster child of the “SaaSpocalypse” thesis. The Journal's Sebastian Herrera talked recently with CEO Marc Benioff about what comes next. Salesforce’s stock is down 28% this year to date

“People think we have our back against the wall when in fact the opportunity has never been greater,” CEO Marc Benioff said in an interview.

By year-end, Salesforce plans to launch a new AI platform that automatically studies users and takes actions on their behalf. An earlier flagship product of that push, Agentforce, has been somewhat slow to gain traction.

Yet there are early signs the bet is working, the Journal reports. Business customers spending big on AI have increased their average median spend on Salesforce in the past three months by 3%, according to data from Andreessen Horowitz. 

WSJLI: I’m curious about your thoughts on Mythos, the model that has spooked the cyber establishment.

Sinha: There used to be 60 days' time when an attacker got in, and before they actually inflicted damage. So between intrusion and breach, the dwell time of the attacker was 60 days. Now it is 27 seconds. And after Mythos, it's near zero... So any hope of detection is pretty much out of the window… you can't prevent the unpreventable. So what Mythos has done, it has accelerated this trend that cyber has to be about resilience, not about prevention.

WSJLI: What do you mean by resilience? 

Sinha: You are fundamentally assuming that attacks and bad things will happen… What would you do first thing? What could I lose? Who are my users? Where is my data? What is the sensitivity of the data? Once you understand that you tighten as many things as possible that are misconfigured or risky… then you think about if it does happen to me, how do I get back into the game? These are the two pieces: risk and remediation.

So now think about Mythos-powered AI attacks. Somebody sitting in North Korea, they decide to rent services on Mythos and create like this new attack software… Now we are talking about machine speed and machine comprehended attacks, which means that your ability to recover or remediate has to be super fast… which means that you need to have preemptive work done.

WSJLI: And what have you heard from customers on the potential threats posed by these new AI models? What are you telling them?

Sinha: My message was consistent… prevention and detection is not sufficient… it means that your resilience strategy has to be even stronger.

WSJLI: Let's stop talking about AI threats and look at the flip side. Rubrik uses Anthropic Cowork and your engineers use Claude Code. How is that working out?

Sinha: Yeah, 100%. And it has accelerated our product development. I started Rubrik 12 years ago and we never had any reviews in Rubrik where somebody said I can do things faster... I did a review of one of our most critical products… and they said they would deliver this product a quarter faster. I couldn't believe it.

It shows that there is a real moment where people are realizing the power of this technology. Because it is not just helping them to do work. It actually is doing the work. And they become the orchestrator or manager of this thing. And one thing that we repeatedly talk about in Rubrik is don't macro manage AI. Let AI cook.

Adding to the pressure on business leaders navigating the technical and financial hurdles of AI deployment is a growing public and political resistance to the technology's rapid expansion.

Verizon Chief Executive Dan Schulman doesn’t pull punches about the pain the technology could unleash on America’s workforce, the Journal reports

“It’s a very difficult time, and everyone knows it is,” Schulman said in an interview. “So I think being authentic, being realistic, telling the truth, as best you can” is key.

Just months into the job, he has predicted 20% to 30% unemployment within the next two to five years.

WATCH | Schulman: ‘We’ll Hit AGI in the Next 2–4 Years. Onstage at the WSJLI Board of Directors Summit in November, Schulman predicted  that human-level intelligence will emerge sooner than most expect.

Skipping the Pollyana-ish AI soundbites.“CEOs are not thinking about this in the right way,” Bill George, the former CEO of Medtronic, tells the Journal. Too many, he said, are focusing only on productivity. “They should be very candid with them and paint the big picture.”

Schulman’s big picture has also included sweeping job cuts. Verizon announced 13,000 layoffs shortly after his appointment as CEO in October. Verizon has said its layoffs weren’t related to AI.

Add tech workers to the growing AI resistance. Even those in AI's hometown of San Francisco say they are uneasy with a growing wealth gap. 

Amazon.com, Meta, Oracle and others are already laying off thousands of people. Many workers are worried that AI might make their programming jobs obsolete.

Here's Jackie Tom, who has been a rentals broker in San Francisco for almost 30 years, talking with the Journal's Tim Higgins about the housing situation in the city.

The unveiling of Mythos AI model has forced both sides to return to the negotiating table, people familiar with the matter tell the Journal.

Axios reports that the National Security Agency is already using Mythos.

AI chip startup Cerebras files for IPO. The company, whose customers include OpenAI and Amazon, didn’t say how many shares it would sell or at what price. A spokesperson said the IPO is set for mid-May.

DeepSeek considers seeking outside money. The Chinese AI startup, creator of that breakout open-source AI model way back in early 2025, is seeking to raise at least $300 million at a valuation of at least $10 billion, the Information reports.

Data center delays. Data center construction in the U.S. is facing significant delays due to labor shortages, permitting obstacles, and energy constraints. Nearly 40% of data center builds are now behind schedule, threatening to delay billions in planned investment returns, the FT reports.

Sam Altman’s side hustles. As questions continue to arise over Altman’s personal investments through Hydrazine, his venture-capital-firm, leading some shareholders to privately question whether he should lead the company through the turbulence of going public, the WSJ reports.

Vice President JD Vance is expected to lead a new round of peace talks with Iran in Pakistan this week in a fresh effort to end the war, but there still appear to be significant gaps between both sides as the U.S. pushes Iran to lock up its nuclear program and reopen the Strait of Hormuz. (WSJ)

Exxon Mobil, Chevron and other energy companies are speeding up their searches for new oil-and-gas prospects—far away from the perils of the war in the Middle East. (WSJ)

A bipartisan group of senators is raising concerns that a merger between United Airlines and American Airlines could harm travelers. (WSJ)

Maine this month joined a growing list of blue states exploring or adopting new taxes on the highest earners, enacting a new 2% surcharge on annual income over $1 million. (WSJ)

The WSJ Tech Council brings together CIOs, CTOs and CISOs advancing innovation and shaping the future. Join this trusted community where tech executives connect with peers to explore emerging trends and gain the perspective they need to stay ahead of disruption.

Request Information

The WSJ CIO Journal Team is Steven Rosenbush, Isabelle Bousquette and Belle Lin.

The editor, Tom Loftus, can be reached at thomas.loftus@wsj.com.