On the surface, Capcom and Virgin Voyages have very little in common. The former is a 46-year-old Japanese video game company firmly centered in the virtual world with popular franchises like “Street Fighter” and “Resident Evil.” The latter is a Florida-based cruise line whose business model is physical, centered on a fleet of four ships.
But both find themselves in Las Vegas this week at a developer conference hosted by Google Cloud, where on Wednesday they will separately debut new applications of AI that do have a common theme: guiding the explorer.
At Capcom, AI agents have been deployed to streamline playtesting for new video game development so that engineers and artists can focus less on debugging and more on creating gaming worlds. Virgin, meanwhile, has unveiled a new AI-enabled virtual assistant called Rovey that can help guests book their trips, offer proactive itinerary recommendations, and answer logistic questions.
Capcom’s AI agents are already operating for over 30,000 hours per month and were built to inspect and pressure-test around half a dozen video game titles before they’re publicly released. “Game teams have become very large, and as these games grow in size and complexity, these game teams find themselves working on very high friction, difficult problems,” says Jack Buser, director of games for Google Cloud.
As these AI agents play through the game, they analyze a checklist of problems that could occur along the way: is the graphics quality failing? Did a crash occur that made the game unplayable? Is there a sense of discomfort as characters move across the field of play? The AI agents aren’t just autonomously detecting and reporting these issues, Capcom says. They are also making suggestions on how to fix them.
“The current game world is as big as one city,” says Kazuki Abe, Capcom’s technical director and head of AI solutions and platform, in an interview conducted with a translator. “There are thousands of characters and tens of thousands of objects, like chairs and desks.” All these variabilities make it difficult for humans to check everything, Abe adds.
For example, visual inspection agents were designed to study when a character changes the equipment they are holding. This simple gesture alone can take 5,280 hours for human playtesters to monitor. AI now screens and flags bugs within those visuals, completing the verification process in about 72 hours. Other agents predict where system issues could occur during the development process. Yet another can assist newer employees, who can ask an AI agent how a veteran engineer would have handled a similar debugging problem in the past.
Over the years, the $200 billion gaming industry has developed more realistic art and graphics and more complex gameplay, resulting in longer and more expensive development cycles for game makers like Capcom. The industry is also facing cyclical challenges. Revenue growth has slowed as consumers spend more time on mobile games. The industry has been beset with layoffs, partly due to consolidation.
Shinichi Inoue, Capcom’s VP of engineering, says agentic AI will help the company train future game designers, artists, and game programmers, a group of employees that the company calls “creators.” “We are using AI to widen the potential of the creators,” Inoue says through a translator. “We are not intending to lower the workforce.”
Virgin’s Rovey, meanwhile, is the cruise company’s first-ever, external-facing AI tool. Booking a cruise, especially for travelers who haven’t done so before, is a complex process involving navigating various trip itineraries, booking restaurant and entertainment reservations, and off-boat excursions.
“We have to use AI to effectively expedite that process,” says Nathan Rosenberg, Virgin’s chief brand and marketing officer, who says he will closely monitor the satisfaction scores for Rovey, as well as how it influences a guest’s likelihood to recommend Virgin to family and friends.
Rovey’s use case emerged from a small working group of AI “champions,” a cross-functional team that looks at how the crew and sales teams perform their jobs and explores new ways to infuse AI into those workflows. The commercial and marketing teams take a fairly active lead in developing Virgin’s AI playbook.
“If you think about most traditional businesses, they’ll tell you first about what the IT team is doing with vendors, or what the IT team is doing to reduce costs,” says Virgin Voyages CEO Nirmal Saverimuttu. “The front door to AI for us is our growth and marketing teams.”
Rovey was thus developed to support both productivity gains and revenue generation. On the ship, any questions Rovey can answer will lessen the burden for the customer service-focused team members. But Rovey is also designed to help speed up conversation. It typically takes Virgin about six to eight weeks to convert a first-time cruise customer. With Rovey, Saverimuttu hopes to reduce that sales cycle to “two to three weeks.”
John Kell
Send thoughts or suggestions to CIO Intelligence here.
Apple CEO Tim Cook steps down. After 15 years of leading the electronics giant, Apple announced that Cook would step down and be succeeded by longtime hardware giant chief John Ternus in September. Cook has frequently been featured on Fortune’s Most Powerful People list and has won praise for driving Apple’s soaring market cap due to explosive sales of the iPhone and for unveiling hit products like the Apple Watch and Apple’s wireless AirPods. And yet, the company has faced some criticism in recent years for making almost no notable advancements when it comes to generative AI. That’s a problem that Ternus will need to address.
Three top executives quit OpenAI in a single day. Kevin Weil, who led the ChatGPT maker’s scientific research arm, announced on Friday that he would be leaving the company as his division was being decentralized into other teams. On the same day, Bill Peebles, who led OpenAI’s shuttered video app Sora, and Srinivas Narayanan, CTO of B2B Applications at OpenAI, also announced their departures. Bloomberg reports that Weil, who previously steered product development teams at Instagram and Twitter before joining OpenAI two years ago, exited because the AI startup is angling to refocus and unify its products. Prism, a scientific workspace program that Weil’s team developed, will become part of the Codex team.
CEOs are getting louder on AI’s job impact. Some of the nation’s most powerful executives are more frequently opining on how AI will change the future of work, including comments from Verizon CEO Dan Schulman, who told the Wall Street Journal that difficult times were ahead “and everyone knows it.” The telecommunications giant set aside $20 million in a career-transition and retraining fund for AI as Verizon began to trim 13,000 jobs last year. Jensen Huang, the CEO of AI chipmaking giant Nvidia, meanwhile remains (somewhat) more bullish, speculating that AI agents will act more like managers than job killers. “Your [AI] agents are harassing you, micromanaging you, and you’re busier than ever,” Huang said during a recent panel discussion. And last week, Tesla and SpaceX CEO Elon Musk called on the federal government to issue checks as the best way to deal with AI-related unemployment.
Anthropic scores $5 billion from Amazon; Mythos generates more buzz. Amazon announced it is investing an additional $5 billion into Anthropic and may put in $20 billion more further down the line, in a deal that also includes the tech giant providing the Claude maker with chips that can reach about 5 gigawatts of power. This comes as Anthropic’s Mythos, the company’s most advanced and still restricted AI model, has banking executives and regulators racing to assess its cybersecurity risks. “On Mythos, look, it’s a serious issue,” Barclays CEO C. S. Venkatakrishnan told Reuters. Top executives from Morgan Stanley, Goldman Sachs, and other U.S. banks have also told investors they are monitoring Mythos’ risks. And the National Security Agency is reportedly using Mythos Preview, despite Anthropic being deemed a “supply chain risk” in its spat with the Department of Defense.
Very few board members are fluent in AI. A mere 23% of governance respondents said their board members had a high level of fluency in AI, while 51% said their basic understanding was “moderate” and 23% said there was low exposure, according to corporate disclosure data and a survey of 130 executives, largely from U.S. public companies, conducted by the nonprofit think tank the Conference Board.
Most of the respondents reported that their boards were building fluency through regular board or committee discussions (67%), internal briefings (49%), and advice from external experts (42%). But far fewer said they were becoming more fluent through more structured approaches including scenario planning or director education courses.
“Lots of independent board directors aren’t necessarily coming from a background that includes generative AI and AI models, and therefore, are still in the early stages of building that fluency,” Andrew Jones, a principal researcher at the Conference Board, tells Fortune. “They’re trying to integrate AI much more into the cadence of the board by getting briefings, but we aren’t seeing any major redesign of the board architecture.”
But perhaps they should accelerate the pace of learning, as the Conference Board notes that the share of large-cap public companies listing AI as a material risk has soared from 12% in 2023 to 83% last year. “Clearly, AI oversight shouldn't just be left to the CTO or the CIO,” says Jones. “It’s a shared and collective responsibility.”
Courtesy of the Conference Board
Hiring:
- W.S. Darley & Co is seeking a chief information/technology officer, based in Itasca, Illinois. Posted salary range: $200K-$300K/year.
- Mead & Hunt is seeking a CTO, based in Madison, Wisconsin. Posted salary range: $270K-$330K/year.
- Logos Space is seeking a head of IT, based in San Diego. Posted salary range: $250K-$310K/year.
- Institute on Aging is seeking a CTO, based in San Francisco. Posted salary range: $295K-$325K/year.
Hired:
- Alight appointed Naveen Baweja as CTO, effective April 29, reporting to CEO Rohit Verma and overseeing the human resources software provider’s technology organization and investments in data and AI. Most recently, Baweja served as VP of technology at entertainment giant Disney. He also previously held leadership roles at CNA Insurance and McKinsey.
- CBIZ promoted Peter Scavuzzo to SVP and chief information and technology officer, succeeding CIO John Fleischer, who retired after 12 years at the accounting firm. Scavuzzo joined CBIZ in 2024 following its acquisition of Marcum LLP, where he served as chief information and digital officer and CEO of Marcum Technology.
- First Commonwealth Financial announced the appointment of Ryan L. Gorney as CIO, effective April 27. Gorney will oversee the Pennsylvania bank’s IT, digital strategy, operations, and enterprise transformation initiatives. He previously served as CIO at Primis Bank and CTO at Panacea Financial. He also held senior leadership roles at KeyBank, Ernst & Young, and Accenture.
- Novus named Olivier Pepin as the media agency’s first-ever CTO, where he will lead the company’s technology, AI, and data engineering. Most recently, Pepin served as chief architect at marketing agency MRM, where he led AI, digital commerce, and marketing technology solutions for clients including General Motors, Johnson & Johnson, and Nestlé.
- Cross Country Healthcare appointed Chris Tyrell as CIO, where he will lead the healthcare staffing and recruitment company’s enterprise technology investment and execution in areas including AI and automation. Tyrell previously served as CTO at warehouse staffing company Eclipse Advantage.
- Sovos appointed David McCann as CTO, joining the tax compliance software provider after most recently serving as CTO at fintech software company Self Financial. McCann also previously held senior technology and product leadership roles at Abrigo and Accruent.
- PCBB announced the appointment of Matthew Wallace as CIO, reporting to CEO Curt Hecker. Wallace will oversee the bank’s technology strategy, modernize core capabilities, and advance data and AI-driven insights. Most recently, he served as CIO at the financial institution the Bancorp.
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。