Tesla deployed a small group of self-driving taxis picking up paying passengers on Sunday in Austin, Texas, with CEO Elon Musk announcing the “robotaxi launch” and social-media influencers posting videos of their first rides.
The event marked the first time Tesla cars without human drivers have carried paying riders, a business that Musk sees as crucial to the electric car maker’s financial future.
He called the moment the “culmination of a decade of hard work” in a post on his social-media platform X and noted that “the AI chip and software teams were built from scratch within Tesla.”
Teslas were spotted early Sunday in a neighborhood called South Congress with no one in the driver’s seat but one person in the passenger seat. The automaker planned a small trial with about 10 vehicles and front-seat riders acting as “safety monitors,” though it remained unclear how much control they had over the vehicles.
In recent days, the automaker sent invites to a select group of influencers for a carefully monitored robotaxi trial in a limited zone. The rides are being offered for a flat fee of $4.20, Musk said on X.
Tesla investor and social-media personality Sawyer Merritt posted videos on X Sunday afternoon showing him ordering, getting picked up and taking a ride to a nearby bar and restaurant, Frazier’s Long and Low, using a Tesla robotaxi app.
If Tesla succeeds with the small deployment, it still faces major challenges in delivering on Musk’s promises to scale up quickly in Austin and other cities, industry experts say.
It could take years or decades for Tesla and self-driving rivals, such as Alphabet’s Waymo, to fully develop a robotaxi industry, said Philip Koopman, a Carnegie Mellon University computer-engineering professor with expertise in autonomous-vehicle technology.
A successful Austin trial for Tesla, he said, would be “the end of the beginning not the beginning of the end.”
Most of Tesla’s sky-high stock value now rests on its ability to deliver robotaxis and humanoid robots, according to many industry analysts. Tesla is by far the world’s most valuable automaker.
As Tesla’s robotaxi-rollout date approached, Texas lawmakers moved to enact autonomous-vehicle rules. Texas Governor Greg Abbott, a Republican, on Friday signed legislation requiring a state permit to operate self-driving vehicles.
The law, which takes effect September 1, signals that state officials from both parties want the driverless-vehicle industry to proceed cautiously.
Tesla did not respond to requests for comment. The governor’s office declined to comment.
“EASY TO GET, EASY TO LOSE”
The law softens the state’s previous anti-regulation stance on autonomous vehicles. A 2017 Texas law specifically prohibited cities from regulating self-driving cars.
The new law requires autonomous-vehicle operators to get approval from the Texas Department of Motor Vehicles before operating on public streets without a human driver. It gives state authorities the power to revoke permits for operators they deem a public danger.
The law also requires firms to provide information on how first responders can deal with their driverless vehicles in emergency situations.
The law’s permit requirements for an “automated motor vehicle” are not onerous but require firms to attest their vehicles can operate legally and safely.
It defines an automated vehicle as having at least “Level 4” autonomous-driving capability under a recognized standard, meaning it can operate with no human driver under specified conditions. Level 5 autonomy is the top level and means a car can drive itself anywhere, under any conditions.
Compliance remains far easier than in some states, notably California, which requires submission of vehicle-testing data under state oversight.
Bryant Walker Smith, a University of South Carolina law professor who focuses on autonomous driving, said it appears any company that meets minimum application requirements will get a Texas permit but could also lose it if problems arise.
“California permits are hard to get, easy to lose,” he said. “In Texas, the permit is easy to get and easy to lose.”
MUSK’S SAFETY PLEDGES
The Tesla robotaxi rollout comes after more than a decade of Musk’s unfulfilled promises to deliver self-driving Teslas.
Musk has said Tesla would be “super paranoid” about robotaxi safety in Austin, including operating in limited areas.
The service in Austin will have other restrictions as well. Tesla plans to avoid bad weather, difficult intersections, and will not carry anyone below age 18.
Commercializing autonomous vehicles has been risky and expensive. GM’s Cruise was shut down after a serious accident. Regulators are closely watching Tesla and its rivals, Waymo and Amazon’s Zoox.
Tesla is also bucking the young industry’s standard practice of relying on multiple technologies to read the road, using only cameras. That, Musk says, will be safe and much less expensive than lidar and radar systems added by rivals.
Norihiko Shirouzu and Abhirup Roy
Buy now, pay later (BNPL) payment options are increasingly popular, particularly among young consumers. A recent survey from J.D. Power shows that 42% of millennials and Gen Z actively use BNPL loans to make purchases.
But BNPL loans have not been incorporated by major credit scoring companies, meaning that lenders attempting to size up borrowers and make lending decisions have had a blind spot regarding these debts.
Until now, that is.
FICO, one of the leading credit scoring companies, announced on Monday that it has launched FICO Score 10 BNPL and FICO Score 10 T BNPL, two new scores that incorporate buy now, pay later data.
Lenders utilizing these scores can now get a more complete picture of a borrowers background, potentially leading to more informed lending decisions. And for borrowers, it may mean a better credit score, especially for those who have responsibly used BNPL services in the past.
We want to make sure that were looking at data that will enable lenders that use the FICO score to have the clearest picture, says Julie May, vice president and general manager of B2B Scores at FICO. This is a product thats been growing rapidly, and we think it’s important to be groundbreaking in this approach.
It can really drive financial inclusion, she adds. Frequently, people use BNPL, and if represented in credit scores, can drive more creditworthiness.
Most scores increase, according to FICO’s research
Earlier this year, FICO published the results of a yearlong analysis done in tandem with BNPL company Affirm. Simulating the impact of BNPL data on FICO scores, the analysis found that Affirm customers with multiple BNPL loans would most likely see their scores increase.
Whats groundbreaking and exciting about these new scores is that theyre going to allow those younger consumers to build credit and build their FICO score, says Ethan Dornhelm, vice president of FICO Scores and Predictive Analytics. For borrowers who have blemished, limited, or no credit history, with the inclusion of BNPL data, borrowers are likely to see a benefit in their scores.
As FICO is preparing to launch the new scoreswhich are expected to be available to lenders later this yearit may mark something of a change in strategy, too.
While many borrowers may have initially been attracted to BNPL loans due to concerns about their own creditworthiness (and potential inability to secure a line a credit, accordingly), BNPL loans may now look more like a credit-enhancing product, rather than an offering of last resort.
One of the early selling points for BNPL firms was that they were not sharing data with the credit bureaus, Ben Danner, senior analyst, credit and commercial at Javelin Strategy & Research, told PaymentsJournal earlier this year. The idea was that this lack of sharing would be attractive to consumers, particularly those concerned with their credit scores.
He continued: However, with potential regulatory changes on the horizon, it makes sense for BNPL vendors to pivot towards a strategy that markets these loans as a credit enhancement tool.
Global markets appeared to take the U.S. strike against nuclear targets in Iran in stride as investors watched Monday to see how Iran will react.The price of oil initially jumped more than 2%, fell and then regained about half that much. U.S. stock futures edged lower and share benchmarks in Europe and Asia also were mostly lower.The attacks on three Iranian sites raised the stakes in the war between Israel and Iran and left questions about what remains of Tehran’s nuclear program. It also increased the possibility that Iran might retaliate, potentially disrupting shipping through the narrow Strait of Hormuz, a waterway through which much of the world’s crude oil passes.The big unknown is what Iran will do, analysts said.The price of Brent crude oil, the international standard, was up 1.2% at $77.91 per barrel. U.S. benchmark crude climbed 1.3% to $74.79.The future for the S&P 500 was little changed, while that for the Dow Jones Industrial Average was down 0.1%. Treasury yields were steady.In Europe, Germany’s DAX lost 0.5% to 23,230.54 and the CAC 40 in Paris fell 0.6% to 7,541.25. Britain’s FTSE 100 shed 0.2% to 8,761.53.Overall, there was no sign of panic.“I believe what we are thinking is or the thinking is that it is going to be a short conflict. The one big hit by the Americans will be effective and then we’ll get back to sort of business as usual, in which case there is no need for an immediate, panicky type of reaction,” said Neil Newman, managing director of Atris Advisory Japan.The conflict began with an Israeli attack against Iran on June 13 that sent oil prices yo-yoing and rattled other markets.Closing off the Strait of Hormuz would be technically difficult but it could severely disrupt transit through it, sending insurance rates spiking and making shippers nervous to move without U.S. Navy escorts. As a major oil producer, Iran may be reluctant to close down the waterway, which is used to transport its own crude, mostly to China. Oil is a major revenue source for the regime.“The situation remains highly fluid, and much hinges on whether Tehran opts for a restrained reaction or a more aggressive course of action,” Kristian Kerr, head of macro strategy at LPL Financial in Charlotte, North Carolina, said in a commentary.Speaking to Fox News on Sunday, U.S. Secretary of State Marco Rubio said disrupting traffic through the strait would be “economic suicide” and would elicit a U.S. response.“I would encourage the Chinese government in Beijing to call them about that because they heavily depend on the Strait of Hormuz for their oil,” Rubio said.When asked about that at a routine briefing in Beijing, Chinese Foreign Ministry spokesperson Guo Jiakun told reporters in Beijing that “China is willing to strengthen communication with Iran and relevant parties to continue playing a constructive role in promoting de-escalation” of the conflict.“The Persian Gulf and its adjacent waters are important international channels for cargo and energy trade. Maintaining security and stability in this region serves the common interests of the international community,” he said.Tom Kloza, chief market analyst at Turner Mason & Co said he expects Iranian leaders to refrain from drastic measures and oil futures to ease back after the initial fears blow over.Disrupting shipping would be “a scorched earth possibility, a Sherman-burning-Atlanta move,” Kloza said.Writing in a report, Ed Yardeni, a long-time analyst, agreed that Tehran leaders would likely hold back.“They aren’t crazy,” he wrote in a note to investors Sunday. “The price of oil should fall and stock markets around the world should climb higher.”Other experts weren’t so sure.Countries are not always rational actors and Tehran could lash out for political or emotional reasons, said Andy Lipow, a Houston analyst who has covered oil markets for 45 years.“If the Strait of Hormuz was completely shut down, oil prices would rise to $120 to $130 a barrel,” Lipow said. That would translate to about $4.50 a gallon at the pump and hurt consumers in other ways, he said.Much of East Asia depends on oil imported through the strait. Taiwan’s Taiex fell 1.4% while the Kospi in South Korea slipped 0.2%.In Tokyo, the Nikkei 225 edged 0.1% lower, with gains for defense contractors, oil companies and miners helping to make up for broad losses.“The U.S. strike on Iran certainly is very good for defense equipment,” Newman of Atris Advisory said, noting that both Japan and South Korea have sizable military manufacturing hubs.Australia’s S&P/ASX fell 0.4%.Hong Kong’s Hang Seng regained lost ground, climbing 0.7%, while the Shanghai Composite index picked up 0.7%.In currency dealings, the U.S. dollar rose to 147.82 Japanese yen from 146.66 yen. The euro fell to $1.1464 from $1.1473.
AP Business Writer Bernard Condon in New York and AP video journalist Mayuko Ono in Tokyo contributed.
Elaine Kurtenbach and Bernard Condon, AP Business Writer
Kroger Co. is joining the list of retailers that plan to scale back on brick-and-mortar locations this year.
The grocery giant said in its first-quarter earnings report on Friday that it will shutter roughly 60 stores over the next 18 months, a move that it expects will provide a “modest financial benefit.”
Kroger said the store closures will not impact its guidance for 2025, during which it expects operating profit of between $4.7 billion and $4.9 billion.
The move to close stores comes about six months after Kroger’s proposed takeover of rival grocery chain Albertsons failed to pass muster with regulators. The $25 billion merger was abandoned, but it was a messy breakup, with both companies ultimately suing each other in attempts to claw back damages.
Kroger’s newly announced closures are likely to attract further criticism of the company in some impacted communities. Although the retailer says all employees affected by the closings will be offered roles at other locations, many cities are already struggling with so-called food deserts, and the loss of even a single grocery store can sometimes create additional hardships for residents.
Which Kroger stores will close?
Kroger has not provided a list of stores that will be impacted. However, local media outlets have reported on a number of locations that have recently closed or will close in the coming months, with some citing local unions or elected officials.
Dickinson, TX
Location is closing in June (Houston Chronicle)
Gassaway, WV
Location is closing in August (WCHS)
The Woodlands, TX
Sterling Ridge location reported closed in May (Community Impact)
McKinney, TX
University Drive location is closing (WFAA)
Dickinson, TX
Location reportedly closing in June (Community Impact)
Abington, VA
Location reportedly closing in September (WFHG)
Bristol, TN
Location reportedly closing in September. (WFHG)
It’s unclear if the above locations are part of the 60 locations included in Kroger’s earnings disclosure on Friday.
Fast Company reached out to Kroger to ask for additional information on which 60 locations will be impacted, including a timeline for when the stores are expected to close. We will update this story if we hear back.
Kroger Co. stock (NYSE: KR) rose almost 10% on Friday following its earnings report.
Cincinnati-based Kroger operates 1,239 grocery stores in 16 states, according to its store locator. It owns a number of well-known regional chains, including Ralphs, QFC, Fred Meyer, and Food 4 Less.
The United Food & Commercial Workers Union, which represents Kroger workers at many stores, did not immediately respond to a request for comment.
This story is developing and may be updated.
Misbehavior on digital platforms can be tricky to manage. Issue warnings, and you risk not deterring bad behavior. Block too readily, and you might drive away your user base and open yourself to accusations of censorship. But a new study, presented at the Conference on Human Factors in Computing Systems, suggests a more effective path forward.
Researchers at Northeastern University and Roblox conducted two large-scale field experiments involving more than 770,000 Roblox users to study how suspension duration affects user behavior. The first experiment compared one-hour versus one-day suspensions for users with a single recent policy violation. The second experiment compared one- to three-day suspensions for users with a second recent violation. They tracked outcomes such as likelihood of reoffending, number of subsequent violations, user reports against offenders, and engagement metrics like days active and total time spent on the platform.
Its very common for platforms to issue consequences for violations of the community standards, but theres actually not a ton of causal evidence in the research about how effective different kinds of consequences are, says Jeffrey Gleason, one of the studys authors and a researcher at Northeastern University and Roblox.
The study found that longer suspensions significantly reduced reoffense rates, the number of consequences, and user reports. Longer suspensions also appeared to make users think twice before misbehaving again, increasing the time it took for them to reoffend.
The longer the suspension, the greater its impact on user behavior. A one-day suspension reduced reoffense rates by 6.7% compared to a one-hour suspension, while a three-day suspension reduced it by 8.1% compared to a one-day suspension. The deterrent effect lasted for at least three weeks, though its impact diminished over time, suggesting that some users eventually reverted to old behaviors.
One promising approach is to address bad behavior early. Harsh penalties for first-time violations were more effective at preventing repeat offenses. A one-day suspension lowered reoffense rates by 12.6% for first-time offenders, compared to just 4.4% for frequent violators. And despite concerns that suspensions might drive users away, the study found that users generally remained active on the platform after being banned.
Theres always been this long debate around what is the trade-off between safety and engagement, and the fact that were seeing that you can do really important safety-related work while not sacrificing engagement is maybe not surprising for us, but may be surprising for other platforms, says Alex Leavitt, principal researcher for trust and safety at Roblox and co-author of the paper.
The experiments only tested suspensions of up to three days, and the effects of longer suspensions remain unknown. Its also uncertain how these findings would translate to other platforms. Still, the research offers valuable insights into curbing bad behavior online.
We now have more public evidence around that fact, says Leavitt, so we can lean into that.
Hello and welcome to Modern CEO! Im Stephanie Mehta, CEO and chief content officer of Mansueto Ventures. Each week this newsletter explores inclusive approaches to leadership drawn from conversations with executives and entrepreneurs, and from the pages of Inc. and Fast Company. If you received this newsletter from a friend, you can sign up to get it yourself every Monday morning.
A new McKinsey & Co. study finds that CEOs and their chief marketing officers (CMOs) are becoming increasingly disconnected, and the gap between CEOs and CMOs perceptions of the marketing function appears to be widening: Nearly two-thirds of chief executives say they are comfortable with modern marketing, up from just half in 2023; meanwhile, only 31% of CMOs believe their CEOs are comfortable in the marketing world, down from four points from a year earlier.
That may be part of the reason CMOs and other creative leaders I met at Cannes Lions last week suggested that chief executives consider spending some time at the annual festival of creativity.
To be sure, most marketing executives arent suggesting CEOs come to Cannes Lions for a deep dive into marketing tactics and metrics. Rather, they feel corporate leaders would benefit from being immersed in an environment that celebrates and inspires groundbreaking multimedia work and its creators.
A creativity immersion experience
I would love to bring CEOs to the basement of the Palais so they could see the work and draw inspiration from that, says Valerie Vargas, senior vice president, content creation and advertising, AT&T, referring to the conference center where Cannes Lions winning entries are exhibited. Vargas also recommends that someone from the head of advertising or marketing hand-pick a series of talks for the CEO to attend, sparking ideas and new ways of thinking about problem-solving. Finally, she suggests the marketing team organize a dinner for the CEO and their agency creative teampeople who usually dont get to meet with the companys top executives.
Others believe a trip to Cannes could help CEOs understand the role that creativity will play as companies adopt generative AI tools and other technologies to increase productivitymaking it harder for many companies to boast an operational or information technology edge. Were moving to a world where creativity is the single most important differentiator among brands, says Zach Kitschke, global chief marketing officer of Canva, the design software company.
David Droga, vice chair of Accenture (and most recently CEO of Accenture Song), goes one step further: Creativity is going to drive the outcomes of AI, he says. Thats what creativity has done every time it has been infused into a new technology. He cites the example of modern photography, a technology that was invented and advanced by chemists and scientists but flourished and took on new energy in the hands of artists and other creatives.
Marisa Thalberg, executive vice president and chief customer and marketing officer for retailer Catalyst Brands, contends that Cannes may not be right for every CEO but says its important for chief executives to understand the value for attendants. For me, Cannes is a uniquely valuable few days, not only for the creative and business inspiration but for the concentrated opportunity for learning, connections, and idea generation, says Thalberg. I can make more happen in a few days that would otherwise take weeks . . . or never happen at all.
Can Cannes bring CEOs and marketers together?
CEOs, are you disconnected from your marketers? Marketers, what dont your chief executives understand about your role? Send your thoughts to me at stephaniemehta@mansueto.com, and Ill try to work the answers into my newsletters.
Watch more: Cannes do
How PepsiCo is using chips to power street food entrepreneurs
Inside Apples award-winning advertising
The future of socializing
Since launching Boardroom in 2019, Rich Kleiman and NBA star Kevin Durant have grown the media company into an influential player that confidently straddles business, sports, and entertainment across content, films, TV, and events. Its newest venture is a membership club that Kleiman sees as key to building a long-lasting brand legacy.
I really want to build a sustainable brand that lasts, Kleiman tells Fast Company. By having this core membership community, and having them become, in a lot of ways, voices of this brand, I thought was really crucial.
Kevin Durant (left) of the Phoenix Suns and Rich Kleiman sit courtside at the 2023 WNBA All-Star Game. [Photo: Ethan Miller/Getty Images]
Launching later this month, the Boardroom Members Club will feature regular members-only events, VIP access to Boardroom flagship activations at NBA All-Star, Art Basel and more, exclusive networking opportunities, and a private digital platform.
For me, I saw this boom in membership clubs in the city, and while they all have their own thing, whether it’s the food or the location or the brand name or the type of people that go there, I didn’t think that there were actually communities there that benefited your career, says Kleiman. And for me, I felt like that was my special sauce, understanding the importance of being in a room with the right mix of people.
Boardrooms media arm churns out newsletters, social posts, and content that reaches over 52 million unique monthly visitors. The company is on track to nearly double revenue in 2025, and average monthly reach has increased by 74% in 2025. Its film and TV output in recent years includes the Apple TV series Swagger, Showtimes Emmy-nominated doc NYC Point Gods, and the Oscar-winning short Two Distant Strangers.
However, it was events like Boardrooms annual CNBC x Boardroom Game Plan Summit that showed Kleiman the potential in combining quality content and the community of people who gather around it. Like an IRL LinkedIn for cool people. I thought that was really exciting, and I wanted to create a version of it that was exclusive to members, he says. I wanted that to feel a bit exclusive, because those conferences can be overwhelming for people that are trying to get information and trying to connect.
Members Club events will have the same vibe and feel of the brands bigger events but with more intimate programming. The big names are still in the room, but make them truly accessible and they understand that like they’re there now to integrate with this community, says Kleiman. And [those big names] want that too. It’s really infectious for anyone at any level, to be around that type of hunger and that type of curiosity and excitement. Seeing our consumers and knowing theyre part of our brand and in our comments and at our parties, but they wanted more, and I wanted to give them more.
The combination of not only connecting with Boardroom content, but with fellow fans and members that can impact their own careers and businesses, is where Kleiman sees the most long-term potential.
For me, the real excitement is creating something that I can point to potentially decades from now and say, That was us, we built that.
Youve no doubt heard of Microsofts Copilot. But can you define exactly what it is?
If you find that question a challenge to answer concisely, youre not alone. In the scramble to convert increasingly ubiquitous generative AI capabilities into recognizably branded tools, Copilot has succeededsometimes to the detriment of its own brand.
Consider the most recent evidence of the Copilot brands cultural profile: a request from the Better Business Bureaus National Advertising Division recently asked Microsoft to adjust its Copilot branding and advertising to avoid confusing or misleading consumers. Part of the problem, it seems, is the way Microsoft has stretched the Copilot brand into a kind of catchall for all things AI. (Microsoft Copilot itself says there are multiple Copilots, each tailored to specific tools and platforms, listing about a half-dozen key examples.)
In part, the NAD argued that Microsofts claims that Copilot boosts productivity and ROI were backed only by a study that actually measured a perception of productivity. It also suggested Microsoft is using the Copilot name across so many AI products and features its not always clear which advertised capabilities apply in which use cases. A description of Copilot working seamlessly across all your data, for example, might mislead a user who wasnt clear on which Copilot-branded tool it referred to.
Based on the context of the claims and universal use of the product description as Copilot, NADs recommendation concluded that consumers would not necessarily understand the differences. A Microsoft statement to Fast Company said that while it disagreed with NADs critical conclusions, it is happy to make small adjustments to help customers better understand the differences between the chat and in-app experiences, or to clarify when a study reflects consumer perception.It’s true that the Copilot name has (to the annoyance of some users) long replicated itself across multiple Microsoft product lines as an all-purpose signifier of AI integration. As The Verge has argued previously, this seems to be partly the fallout from efforts to attract more business customers to pay for Copilot capabilitiesand that plays out in occasionally confusing ways. For example, Microsoft 365 Copilot Chat (formerly Bing Chat Enterprise) is actually distinct from Business Chat for Microsoft 365 Copilot (formerly Business Chat, a component of Teams).
This gets at a deeper issue. The irony may turn out to be that there are just too many Copilots (and not just at Microsoft), and the term is being run into the ground. In fact, it risks becoming less of a signifier than a near-generic cliché (see the sparkle emoji of generative AI branding).
The first significant use of the term copilot as an AI-associated name was actually GitHub Copilot, a 2021 release, which allowed users to work with AI in a form of pair programming, according to a Tech Republic mini history of the term. But Microsoft (which owns GitHub) followed soon after, and swiftly came to dominate its use. Windows keyboards now even include a Copilot keyfeaturing the Copilot logo, a ribbony splotch of colorful gradientsto invoke the Copilot in Windows experience. In practice, Microsofts version of the Copilot brand Security Copilot, Copilot Studio, Copilot in Wordis sometimes a product and sometimes a feature.
All of which has arguably diluted the impact of the Copilot brand. Salesforce, Moodys, and Appian, among other companies, now use copilot in AI-related product names. More generally, in the AI context, copilot has come to serve essentially the function that assistant used to: a humanizing nickname for a variety of tech products that are being sold as not just a digital tool, but a kind of trusted peer.
Of course, one of the challenges facing such products at the moment is living up to outsized promises and hopes, like huge overnight productivity boosts. Oftentimes, these tools feel less like a copilot and more like a temperamental trainee. Thats not to say that generative AI wont deserve the promotion from assistant to something more partner-like. But copilot, as an increasingly common and all-purpose term, can sometimes sound like title inflation. And thats not a brand meaning Microsoft had in mind.
Heres a troubling reality check: We are currently evaluating artificial intelligence in the same way that wed judge a sports car. We act like an AI model is good if it is fast and powerful. But what we really need to assess is whether it makes for a trusted and capable business partner.
The way we approach assessment matters. As AI models begin to play a part in everything from hiring decisions to medical diagnoses, our narrow focus on benchmarks and accuracy rates is creating blind spots that could undermine the very outcomes were trying to achieve. In the long term, it is effectiveness, not efficiency, that matters.
Think about it: When you hire someone for your team, do you only look at their test scores and the speed they work at? Of course not. You consider how they collaborate, whether they share your values, whether they can admit when they dont know something, and how theyll impact your organizations cultureall the things that are critical to strategic success. Yet when it comes to the technology that is increasingly making decisions alongside us, were still stuck on the digital equivalent of standardized test scores.
The Benchmark Trap
Walk into any tech company today, and youll hear executives boasting about their latest performance metrics: Our model achieved 94.7% accuracy! or We reduced token usage by 20%! These numbers sound impressive, but they tell us almost nothing about whether these systems will actually serve human needs effectively.
Despite significant tech advances, evaluation frameworks remain stubbornly focused on performance metrics while largely ignoring ethical, social, and human-centric factors. Its like judging a restaurant solely on how fast it serves food while ignoring whether the meals are nutritious, safe, or actually taste good.
This measurement myopia is leading us astray. Many recent studies have found high levels of bias toward specific demographic groups when AI models are asked to make decisions about individuals in relation to tasks such as hiring, salary recommendations, loan approvals, and sentencing. These outcomes are not just theoretical. For instance, facial recognition systems deployed in law enforcement contexts continue to show higher error rates when identifying people of color. Yet these systems often pass traditional performance tests with flying colors.
The disconnect is stark: Were celebrating technical achievements while peoples lives are being negatively impacted by our measurement blind spots.
Real-World Lessons
IBMs Watson for Oncology was once pitched as a revolutionary breakthrough that would transform cancer care. When measured using traditional metrics, the AI model appeared to be highly impressive, processing vast amounts of medical data rapidly and generating treatment recommendations with clinical sophistication.
However, as Scientific American reported, reality fell far short of this promise. When major cancer centers implemented Watson, significant problems emerged. The systems recommendations often didnt align with best practices, in part because Watson was trained primarily on a limited number of cases from a single institution rather than a comprehensive database of real-world patient outcomes.
The disconnect wasnt in Watsons computational capabilitiesaccording to traditional performance metrics, it functioned as designed. The gap was in its human-centered evaluation capabilities: Did it improve patient outcomes? Did it augment physician expertise effectively? When measured against these standards, Watson struggled to prove its value, leading many healthcare institutions to abandon the system.
Prioritizing dignity
Microsofts Seeing AI is an example of what happens when companies measure success through a human-centered lens from the beginning. As Time magazine reported, the Seeing AI app emerged from Microsofts commitment to accessibility innovation, using computer vision to narrate the visual world for blind and low-vision users.
What sets Seeing AI apart isnt just its technical capabilities but how the development team prioritized human dignity and independence over pure performance metrics. Microsoft worked closely with the blind community throughout the design and testing phases, measuring success not by accuracy percentages alone, but by how effectively the app enhanced the ability of users to navigate their world independently.
This approach created technology that genuinely empowers users, providing real-time audio descriptions that help with everything from selecting groceries to navigating unfamiliar spaces. The lesson: When we start with human outcomes as our primary success metric, we build systems that dont just workthey make life meaningfully better.
Five Critical Dimensions of Success
Smart leaders are moving beyond traditional metrics to evaluate systems across five critical dimensions:
1. Human-AI Collaboration. Rather than measuring performance in isolation, assess how well humans and technology work together. Recent research in the Journal of the American College of Surgeons showed that AI-generated postoperative reports were only half as likely to contain significant discrepancies as those written by surgeons alone. The key insight: a careful division of labor between humans and machines can improve outcomes while leaving humans free to spend more time on what they do best.
2. Ethical Impact and Fairness. Incorporate bias audits and fairness scores as mandatory evaluation metrics. This means continuously assessing whether systems treat all populations equitably and impact human freedom, autonomy, and dignity positively.
3. Stability and Self-Awareness. A Nature Scientific Reports study found performance degradation over time in 91 percent of the models it tested once they were exposed to real-world data. Instead of just measuring a models out-of-the-box accuracy, track performance over time and assess the models ability to identify performance dips and escalate to human oversight when its confidence drops.
4. Value Alignment. As the World Economic Forums 2024 white paper emphasizes, AI models must operate in accordance with core human values if they are to serve humanity effectively. This requires embedding ethical considerations throughout the technology lifecycle.
5. Long-Term Societal Impact Move beyond narrow optimization goals to assess alignment with long-term societal benefits. Consider how technology affects authentic human connections, preserves meaningful work, and serves the broader comunity good.
The Leadership Imperative: Detach and Devote
To transform how your organization measures AI success, embrace the Detach and Devote paradigm we describe in our book TRANSCEND:
Detach from:
Narrow efficiency metrics that ignore human impact
The assumption that replacing human labor is inherently beneficial
Approaches that treat humans as obstacles to optimization
Devote to:
Supporting genuine human connection and collaboration
Preserving meaningful human choice and agency
Serving human needs rather than reshaping humans to serve technological needs
The Path Forward
Forward-thinking leaders implement comprehensive evaluation approaches by starting with the desired human outcomes and then establishing continuous human input loops and measuring results against the goals of human stakeholders.
The companies that get this right wont just build better systemstheyll build more trusted, more valuable, and ultimately more successful businesses. Theyll create technology that doesnt just process data faster but that genuinely enhances human potential and serves societal needs.
The stakes couldnt be higher. As these AI models become more prevalent in critical decisions around hiring, healthcare, criminal justice, and financial services, our measurement approaches will determine whether these models serve humanity well or perpetuate existing inequalities.
In the end, the most important test of all is whether using AI for a task makes human lives genuinely better. The question isnt whether your technology is fast enough but whether its human enough. That is the only metric that ultimately matters.
When Marsya Ancker-Robert was younger, her father used to tell that her that he wanted to be buried naked, under a tree in the woods. The idea horrified Ancker-Robert, but when her father passed away earlier this June, the first call she made was to a Dutch company called Loop Biotech.
Since 2020, Loop Biotech has been making biodegradable caskets out of mycelium, the root-like structure of mushrooms, and hemp. Unlike traditional wooden caskets, which are often treated with chemicals that leech into the soil, the company’s offerings are made of natural materials that enrich the soil as they biodegradea process that only takes 45 days after burial.
So far, Loop Biotech has sold about 2,500 caskets in Europeprimarily in the Netherlands, but also in Germany and other parts of central Europe. But Ancker-Robert’s father, Mark Ancker, has just become the first person in the U.S. to be buried in Loop Biotech’s mycelium casket, called the Living Cocoon. “It was dignified, and beautiful,” says Ancker-Robert, who buried her father in a forest clearing on his property. “I have confidence that my dad will be fully part of the garden by winter.”
[Photo: Loop Biotech]
Growing caskets
Loop Biotech was founded in 2020 by Bob Hendrikx, an architect and biodesigner known for his affinity for nature-based solutions, like a Living Bin that uses sea anemones to “eat” or compost our trash, or a Living Couch that uses algae water to cleanse the air around it. It is part of growing cohort of start-ups shaking up the $622 million green burial market with nature-based solutions. Resting Reef, from London, turns cremated ashes into underwater memorials that double as coral reefs. Coeio, from California, makes burial suits out of mushrooms and other organisms that accelerate decomposition.
For Hendrikx, nature was always the starting point. When the designer first came up with the idea for a mycelium casket, he wasn’t looking for sustainable solutions to burial. He was looking for ways to harness mycelium’s natural ability to recycle dead organic matter into nutrient-rich soil. He knew that mycelium thrives best in soil, which led him to question the kinds of applications it could most benefit. “We call it organism-centered design,” he says, contrasting the approach with human-centered design.
[Photo: Loop Biotech]
Today, Loop caskets are made with mycelium, hemp, and nothing else. The two ingredients are mixed and poured into a mold, and a coffin grows out of that mold in just seven days. But nailing down the exact formula took several years. “Too long,” says Hendrikx with a laugh.
Mycelium is a finicky organism that needs the right conditions to grow and is influenced by several environmental factors such as temperature, humidity, CO2, and oxygen. Even the moon, which influences air pressure on Earth, can have an effect, he says: “Collaborating with nature really allows you to see this interconnectivity of the ecosystem.”
The company has a 1,500-square-meter growing facility in Delft, Netherlands with the capacity to grow 500 caskets at a time. In order to scale, Hendrikx wants to double the production capacity and potentially speed up the growth, too. This facility is what the company calls their blueprint. Once it is optimized, Hendrikx is hoping to replicate the model outside the Netherlands and grow the caskets locally, using local materials.
The company has raised just over $3 million to date, and is planning a new funding round to finance the expansion. And for Hendrikx, the road to success transcends mycelium caskets. “Once we are profitable, we’ve shown the world that you can enrich nature while making money, so it’s a business case for a regenerative business model,” he says.
[Photo: Loop Biotech]
A first for the U.S.
The burial in Maine marks a new chapter in the company’s journey, with mycelium caskets ($3,995) and urns ($395) now available to customers nationwide. But Hendrikx says he’s been getting requests for years.
America is experiencing a green burial revolution. The total number of green burial cemeteries in the U.S. has quadrupled over the past 10 years, going from just over 100 in 2015 to more than 400 by March 2025. Over the past two decades, the nonprofit Green Burial Council has seen a 72% increase in demand from cemeteries for more sustainable end-of-life options.
Loop Biotech’s expansion will likely depend on how willing people are to spend a few extra dollars on a biodegradable casket. An entry-level casket in the United States hovers around $800 for a simple metal burial casket, though average costs range from about $2,000 to $5,000.
It will also depend on how fast the company establishes a robust infrastructure in the country. Already, Hendrikx has built a network of distribution partners and sustainable funeral homes that offer green burial alternatives. They also have a warehouse in Los Angeles where they can ship their products from. But if your local funeral home doesn’t offer mycelium caskets, and you’ve never heard about the company, you may never know it exists.
[Photo: Loop Biotech]
Ancker-Robert found out about Loop Biotech from a Tedx talk that Hendrikx gave in January, and she was really surprised she was the first person to order one. As it happens, Loop Biotech was gearing up to launch in the U.S. on World Environment Day when she called.
Ancker-Robert allowed Hendrikx to film the ceremony, which turned into a small act of generosity for the planet. People made offerings of his favorite foods as well as flowers that Ancker-Robert will plant right above her father’s resting place so they can grow into a perennial flower garden. “The process is helping to turn the grieving process into one of creation and gives me something to daydream about instead of focusing on the loss,” she says. “I would much prefer to think of my father as part of the garden than as a dead body lying in the ground.”
Ancker-Robert describes her father a free spirit who, in the ’80s, would jump into dumpsters to salvage food and drive around his community to distribute it. “There’s a famous picture of the traffic jam on the way to Woodstock. In it there is a young man in a striped shirt sitting on a VW bus looking at the traffic with binoculars,” she says. “That’s my dad.
