 |
|
|
|||||
Three AI companiesOpenAI, Google, and Perplexityare on the verge of receiving approval to sell their technology, hosted on their own cloud systems, directly to the government, a person familiar with the matter tells Fast Company. That authorization will be on a low impact and pilot level, the person said, but constitutes a major step toward independence. That independence could help those companies avoid some of the complications created by ongoing partnerships between AI firms and longtime government tech contractors. As large language models have gone mainstream, AI companies have often relied on tech firms that have already passed arduous government security reviewsincluding Microsoft, Palantir, and Amazon Web Servicesto host their chatbots for federal users. In the early days, these partnerships made it easier for AI labs to quickly get their tech in front of government officials, but also meant ceding at least some control over when and how their AI was made available. The downside of that kind of dependence is now playing out in the brewing feud between Anthropic and the Pentagon, which appears to have been fueled, in part, by its partnership with Palantir. The Defense Department is threatening to cancel a $200 million contract after Anthropic requested limits on the use of its AI for certain applications, including autonomous weapons and mass surveillance. Anthropics Claude model was made available to military officials with the help of Palantir’s systems and was even used in the U.S. operation to capture former Venezuelan President Nicolás Maduro, according to reports. According to Semafor, tensions mounted after an Anthropic official asked a Palantir executive how Claude had been used in the operation, prompting concern inside the Pentagon about the companys willingness to support military applications. This is all to say that not relying on a company like Palantir makes selling to the government far less complicated. In pursuit of that independence, OpenAI, Perplexity, and Google applied for, and received, expedited review of their cloud systems last year under a federal security initiative called FedRAMP 20x. Now, Fast Company can report, theyre almost certain to be approved. These approvals are separate from any decision by a specific federal agency to purchase their products, but they show the companies have taken concrete steps to engage the government on their own terms. Anthropic, by contrast, has leaned heavily on partners like Palantir to help sell its technology to government customers. The company does not appear to have participated in FedRAMP 20x, though its not clear why. Still, the question of independence is one Anthropic has publicly acknowledged. We would also like to be able to directly provide services to governments and not necessarily go through a partner at all times, Michael Sellito, the companys head of global affairs, told FedScoop in 2024. Neither Palantir nor Anthropic responded to Fast Companys request for comment.
Category:
E-Commerce
Sebastian Siemiatkowski, CEO of Swedish fintech company Klarna, says the organization is set to drastically downsize. And he says he shares his outlook on the workforce with another CEO: Anthropics Dario Amodei. Siemiatkowski made the comments on the 20 VC podcast with Harry Stebbings earlier this week, where the CEO didn’t deny that the company has been steadily shrinking.The CEO said that currently the company has about 3,000 employees. That’s down from 7,000 just four years ago. In another four, he says there will likely be less than 2,000a reduction of one-third. Siemiatkowski cited both layoffs and the employees leaving the company and not being replaced, and explained that AI’s integration allows for fewer employees.Klarna’s slimming down comes even as buy now, pay later (BNPL) services are booming. Around 30% of Americans say they have used them, according to a 2025 Bankrate report. And in 2025, according to a PartnerCentric survey, 35% said they planned to use the services even more. The popularity is driven by the fact that Klarna, like other BNPL options, allows shoppers to split purchases into interest-free installments, pay within 30 days, or even opt for longer-term financing options. Likewise, thousands of retailers now accept BNPL. Still, the success of such businesses no longer seems to equate to the need for more employees, as AI’s impact looms largersomething some leaders have been increasingly warning about. Anthropic’s CEO Dario Amodei wrote of his gravest concerns about AI in a recent essay that included items like loss of autonomy, “misuse for destruction” and “powerful AI” which he writes is “definitely coming.” Amodei writes: “I think it should be clear that this is a dangerous situationa report from a competent national security official to a head of state would probably contain words like ‘the single most serious national security threat weve faced in a century, possibly ever.’ It seems like something the best minds of civilization should be focused on.” The CEO also predicted AI could cut 50 percent of all white-collar entry-level jobs in the next one to five years, doubling down on a stance he’s warned about previously. Worryingly, Klarna’s CEO doesn’t disagree with Amodei’s stance, acknowledging that he’s “in Dario’s camp” on concerns around AI. “I want to be honest about the fact that I do think there’s going to be a very big shift,” Siemiatkowski said on the podcast. Specifically, he echoed the concerns around job loss. “I’m an optimist at heart, but I also want to be a realist around what’s going to happen in the shorter term, and it’s going to be a lot of turmoil in this.” Regardless, while the CEO seemed to express some major concerns around AI’s rapid advancements, Siemiatkowski has leaned into them heavily. In 2024, he announced that AI could handle a growing number of jobs as the company paused hiring and cut 2,000 employees. But it wasn’t long before customer satisfaction dipped, and the company had to scramble to reassign workers to customer support to handle the fallout. The CEO later took to X to explain what went wrong, writing that he was “tremendously embarrassed” about the turn of events. Fast Company reached out to Klarna to inquire on whether the company would scale back its relationship with AI. A representative said Klarna “did not lean too much into AI,” but its “thinking on human customer service” has changed. The representative continued, “When you automate a large amount of the simpler customer service requests, you are left with the most complex and sensitive cases . . . So we have begun to directly hire a small number of human agents directly employed by Klarna, not at outsourced providers.”
Category:
E-Commerce
When I cofounded Brilliant Earth in 2005, e-commerce was still in its infancy. I believed technology could reshape the jewelry industry entirelychanging how customers find pieces they love, personalizing their own designs, and reimagining the customer experience. We launched as a digital-first venture to do just that. Now, two decades into our pioneering digital journey, I’ve realized something surprising: Our most sophisticated online tools have actually made in-person interactions more valuable. I believe the brands leading the next wave of innovation aren’t choosing between digital and physical. They’re using digital excellence to help create meaningful in-person connections and lifelong brand affinity. BOLSTER HUMAN CONNECTION THROUGH DIGITAL Over the years, Brilliant Earth introduced several industry firsts. Before we launched, buying diamond jewelry online was unheard of; we became one of the first companies to sell engagement rings and fine jewelry through e-commerce, and were early to offer lab-grown diamonds when few knew what they were. We developed online tools for customers to design their own engagement rings and fine jewelry. But we also understood that the future wouldnt be dominated by digital. Seamlessly integrating online and physical experiences would be key. Even as digital tools get more sophisticated, leading companies aren’t eliminating the in-person element. They’re doing the opposite: harnessing these innovations to make human touchpoints more meaningful. In doing so, they create branded experiences that feel more personal and authentic than traditional retail was doing. This transformation spans every retail categoryfrom IKEA’s Place app that lets customers visualize furniture in their homes to Nike’s data-driven inventory localization, based on neighborhood shopping patterns. Even legacy media companies are getting involved: Condé Nast is preparing to launch Vette, a creator-commerce platform using AI-driven tools to help influencers curate personalized storefronts. It bridges the gap between content discovery and purchase decisions. The key insight for any industry leader is this: Modern shoppers no longer see a divide between online and offline shopping. Both are now intrinsic parts of the customer journey, not opposing concepts. No matter your industry, customers expect each digital tool to serve a dual purpose: enhancing their online experience while creating richer, more meaningful interaction when they choose to engage in person. The companies getting this right aren’t using digital tools to replace human touchpoints. They’re giving their teams better information, so every interaction feels more personal and valuable. 3 PILLARS OF NEXT-GENERATION EXPERIENCE As I look toward the future, I see three fundamental principles driving the evolution of exceptional customer experiences: 1. Invest in tech-powered personalization: Digital tools that recognize customers across all touchpoints are delivering increasingly sophisticated personalization. Take Ralph Lauren’s “Ask Ralph,” which launched last fall. Its like having a personal stylist in your pocketone who knows the brands entire archive, drawing upon that knowledge to provide personalized styling advice, complete with live inventory. The experience feels like a natural extension of the world Ralph has built for almost 60 years. In the big box world, Walmart is using AI features like a personalized algorithm that can predict and preempt customer needs, including frequency and quantity of orders. It expanded testing of a GenAI shopping assistant to help customers make the best choices for their needs. At Brilliant Earth, we’ve taken this approach into our showrooms where we use dozens of customer data points to curate each appointment and continue that personalized journey long after customers leave our stores. But we’ve also learned to lean into what our customers find most special: the unique guidance and expertise they get from our jewelry specialists. The goal is to build unified systems that remember customer preferences no matter how they shop with you, so every interaction feels personal and connected. 2. Focus on immersive technologies to eliminate friction: Deploying smart technologies to allow customers to experience products before purchasing will fundamentally change how customers interact with your brand world. Released last summer, Warby Parker’s new AI-powered Advisor feature scans customers’ faces to capture style preferences and measurements, then recommends glasses while offering virtual try-on capabilities that calculate frame sizes. And Sephora launched its latest digital extension this fall called My Sephora Storefront, where creators can build their own beauty shops inside Sephora’s website and app, so shoppers never have to leave the platform. These integrations eliminate the traditional friction points between discovery and purchase, and allow customers to experience products virtually before committing. 3. Emphasize human-AI collaboration: The most successful AI implementations preserve authentic human connections while augmenting capabilities. Condé Nast’s Vette platform demonstrates another approach by using AI to handle complex backend operationslike inventory management, product recommendations, and sales analytics. This allows creators to focus on what humans do best: building authentic relationships with their audiences and providing personal curation reflecting their unique voice and taste. The best AI implementations don’t replace peoplethey make them better at their jobs. Smart companies are learning to use AI to give their teams insights about individual customers, making every interaction more relevant without losing the human touch that builds real loyalty. THE NEXT WAVE OF INNOVATION I’m proud of how far we’ve come in using innovation to enable self-expression and connect with authentic human needs at Brilliant Earth. But more than that, I’m energized by the opportunities ahead. Todays digital transformation has given us capabilities that seemed impossible just a few years ago. While there are many questions about AI’s future impact on jobs and human connection to be answered, the companies and leaders succeeding right now aren’t using AI to eliminate human contact. They’re using it to make those human moments count even more. Those who get this right will reshape the next generation of retail. Beth Gerstein is the cofounder and CEO of Brilliant Earth.
Category:
E-Commerce
Mark Zuckerberg and opposing lawyers dueled in a Los Angeles courtroom on Wednesday, where the Meta CEO answered questions about young peoples use of Instagram, his congressional testimony, and internal advice hes received about being authentic and not robotic. Zuckerberg’s testimony is part of an unprecedented social media trial that questions whether Meta’s platforms deliberately addict and harm children. Attorneys representing the plaintiff, a now 20-year-old woman identified by the initials KGM, claim her early use of social media addicted her to the technology and exacerbated depression and suicidal thoughts. Meta Platforms and Googles YouTube are the two remaining defendants in the case, which TikTok and Snap have settled. Beginning his questioning, the plaintiff’s attorney Mark Lanier laid out three options of what people can do regarding vulnerable people: help them, ignore them, or prey upon them and use them for our own ends. Zuckerberg said he agrees the last option is not what a reasonable company should do, saying, I think a reasonable company should try to help the people that use its services. When he was asked about his compensation, Zuckerberg said he has pledged to give almost all of his money to charity, focusing on scientific research. Lanier asked him how much money he has pledged to victims impacted by social media, to which Zuckerberg replied, I disagree with the characterization of your question. Lanier also asked Zuckerberg about what he characterized as extensive media training, including for testimonies like the one he was giving in court. Lanier pointed to an internal document about feedback on Zuckerberg’s tone of voice on his own social media, imploring him to come off as authentic, direct, human, insightful and real, and instructing him to not try hard, fake, robotic, corporate or cheesy in his communication. Zuckerberg pushed back against the idea that hes been coached on how to respond to questions or present himself, saying those offering the advice were just giving feedback. Regarding his media appearances and public speaking, Zuckerberg said, I think Im actually well known to be sort of bad at this. The Meta CEO has long been mocked online for appearing robotic and, when he was younger, nervous when speaking publicly. In 2010, during an interview with renowned tech journalists Kara Swisher and Walt Mossberg, he was sweating so profusely that Swisher asked him if he wanted to take off the hoodie that was his uniform at the time. Lanier spent a considerable stretch of his limited time with Zuckerberg asking about the companys age verification policies. I dont see why this is so complicated, Zuckerberg said after a lengthy back-and-forth, reiterating that the companys policy restricts users under the age of 13 and that they work to detect users who have lied about their ages to bypass restrictions. Zuckerberg mostly stuck to his talking points, referencing his goal of building a platform that is valuable to users and, on multiple occasions, saying he disagreed with Laniers characterization of his questions or of Zuckerbergs own comments. Zuckerberg has testified in other trials and answered questions from Congress about youth safety on Meta’s platforms, and he apologized to families at that hearing whose lives had been upended by tragedies they believed were because of social media. This trial, though, marks the first time Zuckerberg will answer similar questions in front of a jury. And, again, bereaved parents are expected to be in the limited courtroom seats available to the public. The case, along with two others, has been selected as a bellwether trial, meaning its outcome could impact how thousands of similar lawsuits against social media companies are likely to play out. A Meta spokesperson said the company strongly disagrees with the allegations in the lawsuit and said they are confident the evidence will show our longstanding commitment to supporting young people. One of Meta’s attorneys, Paul Schmidt, said in his opening statement that the company is not disputing that KGM experienced mental health struggles, but rather disputing that Instagram played a substantial factor in those struggles. He pointed to medical records that showed a turbulent home life, and both he and an attorney representing YouTube argue she turned to their platforms as a coping mechanism or a means of escaping her mental health struggles. Zuckerberg’s testimony comes a week after that of Adam Mosseri, the head of Meta’s Instagram, who said in the courtroom that he disagrees with the idea that people can be clinically addicted to social media platforms. Mosseri maintained that Instagram works hard to protect young people using the service, and said it’s not good for the company, over the long run, to make decisions that profit for us but are poor for peoples well-being. Much of Mosseri’s questioning from the plaintiff’s lawyer, Mark Lanier, centered on cosmetic filters on Instagram that changed peoples appearance a topic that Lanier is sure to revisit with Zuckerberg. He is also expected to face questions about Instagrams algorithm, the infinite nature of Metas feeds and other features the plaintiffs argue are designed to get users hooked. Meta is also facing a separate trial in New Mexico that began last week. Kaitlyn Huamani and Barbara Ortutay, AP technology writers
Category:
E-Commerce
A new $7.25 billion settlement between Bayer and a group of cancer patients could wrap up a huge wave of lawsuits against the company over allegations that it didnt warn consumers about cancer risks associated with the weedkiller Roundup. Bayer faces more than 180,000 claims over Roundup, which contains the herbicide glyphosate the chemical at the center of the controversy. Most of those claims are from people who used the weedkiller, which is sold at any hardware or garden store, at home. The lawsuits have prompted Bayer to pull glyphosate out of many products under the Roundup brand, though glyphosate is still commonly used by farmers and in the agriculture business broadly. The science around glyphosate is controversial. The Environmental Protection Agency has said that glyphosate is not likely to cause cancer in users if applied as directed, and does not require companies selling it to include a warning about links to cancer. The World Health Organizations cancer agency classified the chemical as a substance likely to cause cancer in humans more than a decade ago, though those findings faced scrutiny a few years later over reports that the published version differed from a draft. Just last month, a landmark study determining that glyphosate didnt pose a risk to human health was retracted, 25 years after its publication. The retraction, prompted by emails revealing Monsantos influence, undermines a longstanding regulatory foundation that has cited the key research for decades. A long battle The Supreme Court is set to hear arguments over Bayers effort to fend off an onslaught of cancer-related lawsuits over Roundup in April. While the new settlement proposal wont affect that case, it could help both Bayer and the plaintiffs hedge their bets if the Supreme Court doesnt side in their favor. Bayer, a German pharmaceutical and biotech giant, is best known for making the common pain reliever Aspirin. The company acquired Roundup maker Monsanto in 2018 for $63 billion, betting that owning a major player in the agriculture business would diversify its business and pay dividends down the road as farming supplies boomed. That hasnt come to pass, and Monsantos costly litigation has further dragged Bayers share price down from its highs around a decade ago. Today, Bayer is worth less than the price it once paid for Monsanto. Hundreds of thousands of lawsuits Out of the cases against Bayer over Roundup so far, only a sliver were decided by a jury, yielding 13 decisions favoring the pharmaceutical giant and 11 siding with plaintiffs. Last year, a jury in Georgia ordered Bayer to pay $2.1 billion in damages to a plaintiff who suffers from non-Hodgkins lymphoma, a cancer that begins in white blood cells. Some other cases have been resolved in separate settlements, but many remain unresolved. Under the terms of the proposed settlement, Bayer would make payments into a designated fund on a yearly basis for 21 years, which could total up to $7.25 billion, to resolve most of the outstanding Roundup lawsuits. That money would then be doled out to people based on their Roundup usage, age of cancer diagnosis and the severity of their disease. Under the settlements terms, agricultural and industrial workers who faced regular exposure to the products chemicals could receive an average of $165,000 if they were diagnosed with non-Hodgkin lymphoma under the age of 60. Residential users, those diagnosed later in life, and those with less aggressive cancer could receive tens of thousands in compensation. Bayer has said that it could still cancel the settlement, which does not yet have court approval, if too many plaintiffs decide to opt out and reject its terms. Supreme Court poised to decide Last month, the Supreme Court said that it would hear a case on the issue in order to determine if federal laws protect Bayer, which complies with the Environmental Protection Agencys rules, from lawsuits filed in state courts. The EPA does not require products including glyphosate to be sold with a cancer warning. Bayer praised the Supreme Courts decision to take the case, arguing that farmers need regulatory clarity around the widely used product and calling the milestone an important part of its effort to significantly contain litigation around Roundup. It is time for the U.S. legal system to establish that companies should not be punished under state laws for complying with federal warning label requirements, Bayer CEO Bill Anderson said.
Category:
E-Commerce