Wednesday, May 6, 2026

How Wyndham scales AI to improve hospitality at 8,400 hotels

The business model for Wyndham Hotels & Resorts purposefully embraces sprawl. The hotel franchising company has about 8,400 hotels across six continents, comprising a portfolio of 25 brands, including Days Inn, La Quinta, and Ramada.

But the $450 million that Wyndham has spent on technology since 2018 favors standardization and fewer vendors. “We realized years ago that it was better for our franchisees if we created an overall technology bundle,” says Scott Strickland, Wyndham’s chief commercial officer. “We have buying power because we represent thousands of hotels that allow us to get things at a lower price.”

A former CIO at three organizations—Black & Decker, D&M Holdings, and Wyndham—Strickland became CCO in January 2024 to oversee all the traditional CIO responsibilities, including IT and cybersecurity, but also oversee revenue management decisions like pricing, marketing, and public relations.

He has been the central figure steering Wyndham’s decade-long technology overhaul that includes lifting and shifting to the cloud, hiring and building out a centralized data team, and mandating that all properties use a centralized reservation system built by Aven Hospitality, a software-as-a-service vendor that specializes in the hospitality sector. Strickland says that these infrastructure investments have also made it easier for Wyndham to deploy generative AI features.

To get feedback from Wyndham’s base of hotel property owners, Strickland formed a franchisee technology committee to learn more about their biggest technology and business priorities. They came back with a list of 72 requests, such as adding EV charging stations. Several concepts were consolidated and led to the creation of the AI-enabled guest engagement platform, Wyndham Connect, which uses AI to send proactive texts or respond to inbound guest questions, handle mobile check-ins, and allow staff to use AI-generated messaging to answer questions.

“What it does for the guests is it removes friction from the entire journey,” says Strickland.

Jason Hoehne, general manager of the 53-room AmericInn by Wyndham hotel in Lincoln, Nebraska, was among the pilot participants for Wyndham Connect. The hotel owner says he’s used the tool to upsell, speed up the check-in process, more efficiently track when guests depart and a room is ready for housekeeping, and add more layers of communication with guests, but led by AI.

“We would never replace a front desk agent with a kiosk or device, that’s not my ownership style,” says Hoehne. “But it gives the staff tools to be more efficient and productive.”

He adds that the system both adds incremental revenue and boosts productivity. On the former, hotel owners can use the tool to upsell early check-ins or extended departures, while mobile-enabled digital tipping has led to an average tip of close to $10 for housekeepers. The latter can help incentivize staff to stay at Wyndham properties, lowering costly turnover rates.

Unlike other technology offerings from the corporate parent, Wyndham Connect and the AI functionality isn’t mandated. Still, the company said the tool has been embraced by over 5,000 hotels across North America and 100 properties internationally to handle 4.6 million mobile check-ins, $9 million in upselling, 12 million AI messaging engagements, and a 300-basis-points increase in direct booking for hotels using the AI voice functionality.

More recently, Wyndham rolled out an AI feature that enables AI to handle calls to the front desk. Around 1,100 hotels are using that tool today.

On Wednesday, Wyndham also rolled out its newest travel app on ChatGPT, with the company claiming it is the first hotelier to have direct integrations with the three largest LLMs (Google’s Gemini and Anthropic’s Claude are the other two). Wyndham says it is embracing AI in this manner because conversational and visual travel discovery is migrating to these chatbots, with Strickland citing the scale of ChatGPT’s weekly user base of 900 million.

“It needs structured data to understand things about your hotel,” says Strickland. “It can get that data by scraping your website, but it can’t get everything it needs to execute a booking. We created an app that has all that data that it needs to help someone through that booking.”

For internal applications of AI, Strickland has leaned on existing vendors such as Google, Amazon Web Services, and Salesforce’s autonomous AI agents platform, Agentforce. “Go ahead and use the agent that’s closest to the data source,” says Strickland. “We are multimodal, and we’re also using OpenAI for our internal, large LLM. We’re going to look at every use case and apply the best one.”

As for employee training, Strickland says Wyndham offers courses and has leaned on vendors like Amazon and Atlassian.

Wyndham’s agentic AI deployments include allowing customers to book a room directly through an agent. For those who prefer to speak to a staff member, that option is still available, but also with AI in a less obvious way. Strickland says AI assists employees with the booking process, including sharing proactive information like if a caller is a loyalty member or perhaps complained about noise during their last stay and should be offered a quieter room. When used in this manner, AI can lead to better hospitality, Strickland says.

“If you’re the person that just wants the AI and goes straight to your room, we have it,” says Strickland. “If you want to talk to a front desk agent, we have it. But we’re going to make that engagement now even better than it used to be, with AI.”

John Kell

Send thoughts or suggestions to CIO Intelligence here.

NEWS PACKETS

Big Tech’s big earnings and bigger AI spending. Overall, the largest technology firms reported strong quarterly earnings last week, but as Wall Street likes to do, it picked winners and losers when taking a closer look at the numbers. Google had a particularly strong quarter that sent shares to an all-time high, due to strong digital ad sales for its search engine and fast growth from its Cloud division. Shares at Meta, conversely, were battered after the Instagram and Facebook parent reported a drop of 20 million users, big AI spending, and that it continues to pour billions into its money-losing metaverse and virtual reality business. But the collective figure that Wall Street is keeping a particularly close eye on is AI capital expenditures: they could surpass $700 billion in 2026, and some analysts expect they will climb above $1 trillion in 2027.

Washington also has a big tech week. President Donald Trump’s light-touch approach to AI regulation appears to be evolving, as the New York Times reports that the federal government is weighing an executive order that would explore potential oversight of new AI models before they are publicly released. Shortly after that story was published, Bloomberg reported that Google, Microsoft, and xAI had joined OpenAI and Anthropic in agreeing to give the U.S. government early access to their AI models, which would be reviewed by the Commerce Department’s Center for AI Standards and Innovation. In other related news, the White House’s fraught relationship with Anthropic continues as it opposed the AI startup’s plan to expand access to its Mythos model, while separately, the Pentagon has inked agreements with Nvidia, Microsoft, Reflection AI, and Amazon to expand the use of AI tools on classified military networks.

PayPal, Coinbase latest to cut jobs amid AI push. The Wall Street Journal reports that PayPal, which is in the process of reorganization operations to make the mobile payments app Venmo a standalone unit, plans to cut 20% of its staff over the next two to three years as it accelerates its adoption of AI and cuts costs. Coinbase, meanwhile, on Tuesday announced it would trim 14% of its staff, partly due to a crypto downturn but also a reflection of AI. Fortune reports that the company’s leadership structure will become flatter and cut managers in favor of “player-coaches,” who oversee team members but can also be individual contributors. The more AI savvy employees may also be leveraged to direct agents in one-person teams that would include the responsibilities of engineers and designers.

Elon Musk goes to trial against OpenAI. The legal squabble between the billionaire founder of SpaceX and the ChatGPT maker took a few twists in federal court over the past week, in a case that centers on Musk’s complaint that OpenAI and CEO Sam Altman breached the AI startup’s founding mission as a nonprofit by becoming a commercial enterprise. Musk himself took to the stand, at times facing terse cross-examination from OpenAI’s lawyers, making the case that he was “a fool who provided them free funding to create a startup.” Interestingly, two days before the trial was to start, Musk reached out to Greg Brockman, OpenAI’s president and co-founder, to ask him if he wanted to settle the case. OpenAI has consistently said Musk’s claims are “baseless.”

Study shows OpenAI’s model outperforms doctors. A Harvard study comparing the diagnosis of 76 cases at a Boston hospital found that an AI reasoning model developed by OpenAI outperformed human clinicians and prior model generations. The research, which was published in the journal Science, centered on Open AI’s model and two physicians at three diagnostic touch points, beginning in the ER through admission to the hospital, and found that AI identified the exact or very close diagnosis in 67% of the cases, while human doctors were right 50% to 55% of the time. Already, more than 80% of physicians are using AI tools, according to a recent survey published by the American Medical Association, mostly for lower stakes use cases like combing through medical research and documentation. But the Harvard research points to a future where AI may play more of a role in clinical settings.

ADOPTION CURVE

AI agents are outpacing security guardrails. 86% of organizations say they expect AI agents will outpace the security guardrails they have in place within the next 12 months, with more than one out of every two saying it will occur in the next six months, according to a survey of more than 1,600 IT and security leaders conducted by cybersecurity vendor Rubrik.

The study also found that only 23% of technologists report having full visibility into the AI agents that are operating within their enterprises. Nearly nine out of ten (88%) say they lack the ability to roll back agent actions without disrupting their systems.

Kavitha Mariappan, Rubrik's chief transformation officer, tells Fortune that AI agents have “the potential to operate outside of IT’s view if the right sort of guardrails are not there.” While many CIOs have placed guardrails around data and tasks that AI agents can perform, Mariappan says they need to extend their thinking beyond visibility into what data is accessed and the outputs that are being produced and be more focused on “observability,” which encompasses context, intent, and reasoning.

“Real-time guardrails need to be implemented,” says Mariappan. “It goes beyond identity and permissions. We're in that world now where we're talking about business logic that is applied in real time.”

Courtesy of Rubrik