Discussion | Link

A Silicon Valley software maker and an ecommerce company reveal to WIRED how they are navigating the emerging challenge of “tokenomics.”

Discussion | Link

Discussion | Link

Tony Latham. Bacardi North America This as-told-to essay is based on a conversation with Tony Latham, the 53-year-old President of Bacardi North America, based in Miami. It's been edited for length and clarity. I'm from Australia, and I've worked in consumer goods at well-known companies such as SC Johnson, Danone, and Unilever. The US is the eighth country I've lived in in the past 25 years; I moved around a lot with those companies. I spent about 10 years in Asia, about eight in Europe, and when I joined Bacardi in 2019, I was the global CFO, based in Bermuda. Business Insider's Power Hours series gives readers an inside look at how powerful leaders in business structure their workday. Reach out to editor Lauryn Haas to share your daily routine. I took over as president of Bacardi North America in April 2023 and moved to Miami. Here's a typical day in my life. I wake up at 5:30 a.m. I spent some time in the military, so I like getting up early. I enjoy a bit of peace and solitude by myself in the morning. I have an Oura ring, and I was an early adopter of wearable fitness technology. I always check the quality of my sleep and my readiness for the day, and make any adjustments I feel I need based on the technology's inputs. In the first 30 minutes of my day, I have an espresso from my Nespresso machine I also have a banana and some yogurt before everyone else in the house wakes up, which is quite nice. I like to work out four or five days a week, and then a couple of times on the weekend. I have a three-car garage that I've converted into a home gym. I live with my wife of 20 years, who's Canadian, and our two sons — one is a high-school junior, and the other just graduated. After working out, I have a second breakfast By that time, the house is pretty hectic with the boys waking up and rushing out the door. I tend to have a second breakfast on the commute to work, or when I get to the office. I've got about a 45-minute commute, and that gives me an opportunity to either make some calls (our headquarters in Bermuda is an hour ahead of us) or reflect on the day ahead and listen to some music. Whenever I'm in Miami, I'm in the office. We've recently opened a new one, and it's a beautiful, modern space in the Coral Gables area. I travel anywhere from two to three weeks a month Depending on meetings in Bermuda, I have a lot of domestic travel, visiting our key markets, our commercial people out in the field, and customers. There could be brand activations around big sporting events or big music festivals, where I may be taking customers or interacting with consumers. I try not to schedule meetings before 10 a.m. I'm not always successful at that. When I get to the office, I like to get through a few things early in the morning and set myself up for a day where I can be fully present for the meetings I'm in. Once my meetings start, they're scheduled through until about 3 p.m. I use AI in my workday to collect and analyze data across the business. Having information at our fingertips helps us make decisions more quickly. When considering potential personal investments, I love using AI to generate summaries. I might ask an AI for a quick moat assessment to understand a company's competitive advantages before I invest. I have lunch at the office We have lunch catered in the office. I try to meet for lunch in our common space. There's usually a lot of conversation and banter. It's a great opportunity to recharge in the middle of the day for the afternoon. I like to be visible and available to the team. I spend a lot of time walking around the office and checking in with people. After lunch, it's back to meetings I've got visibility on my schedule two months out, but there can be some fluidity around travel. Bacardi's global leadership team meetings in Bermuda and our board meetings are scheduled more than 12 months in advance. At 3 p.m., after meetings are over, I might go to a local account to provide support. It could be an on-premises venue, a bar, or a restaurant. I meet with the owners before they get busy. If I'm still at the office, I typically leave between 5 p.m. and 6 p.m. The traffic in Miami has gotten increasingly worse over the past couple of years. I'll quite often have a look at the traffic and make a decision based on the traffic patterns. My boys both play sports, so I might be running back to pick one of them up from training or to get them to a sporting match. On the first three days of the week, my wife cooks dinner I have business dinners a few times a week. The household at the end of the day is a bit manic. My wife is a Pilates instructor, so she might be out teaching courses between 6 p.m. and 8 p.m. If we get to sit down and have a meal together at the house, that's great. Occasionally, we'll order in, particularly as the week comes to an end and everybody might be getting a little weary. I like to cook on the weekends. I love to grill, and I've got a really good outdoor kitchen set up. I like to make homemade pizzas and steaks on the weekends. Miami is a great place to visit, so we often have visitors, particularly in the winter months, from up north. My wife's parents sometimes look to escape the Canadian winter and come down to spend some winter months here. We love hosting and entertaining. I try to avoid going back to work after dinner I find that my productivity levels are not what they were at the beginning of the day. In the evenings, I like to read about current affairs. I've gotten into streaming a few shows, though I'm not one to sit in front of the TV for long. When I travel, I work on the plane and stream a few shows. I've been watching The Night Agent, The Recruit, and The Gentleman. I try to get to bed around 9:30 p.m. I find that sleep is by far the most important thing for me right now, so I try to eat a little earlier. I aim for a good seven-plus hours of high-quality sleep, but I'm all about a balanced lifestyle. I love making a cocktail, particularly when I'm barbecuing on the weekend. Among my go-to cocktails is the Bacardi Ocho Old Fashioned, where the aged rum complements the citrus and aromatic bitters. As for my favorite Bacardi product, if I had to pick now, I'm really enjoying Patrón Reposado, which adds richness and a hint of sweetness to a spicy margarita. If I have a cocktail, I keep it to one or two drinks, drink plenty of water, and try not to have them too late in the evening. I try to keep the getting up early routine going on the weekends I use that time to do a bit of work, trying not to disrupt the day's balance while still keeping that routine of getting up. I eat something, do a bit of work, and work out because the boys tend to sleep in a little longer on weekends, which gives my wife and me time to connect. Read the original article on Business Insider

David Selinger. Courtesy of David Selinger David Selinger cofounded Deep Sentinel, an AI security firm, after working at Amazon in the early 2000s. At Amazon, Selinger worked directly with Jeff Bezos, which shaped his AI and business acumen. Deep Sentinel, backed by Bezos, uses AI to prevent crime and innovate home security. This as-told-to essay is based on a conversation with David Selinger, the 48-year-old CEO of Deep Sentinel, based in Pleasanton, California. It's been edited for length and clarity. I went to Stanford but didn't finish my degree, so I went back in 2000 to finish. In 2002, I got hired at Amazon through on-campus interviews while still in college. I worked at Dutch Bros, now a publicly traded coffee company, then. I was running their tech side. I didn't initially want to work at Amazon, but I considered it because I was bored. I had a great job at Dutch Bros, worked very few hours, and made a ton of money. I took the Amazon interview very nonchalantly. I found the data side of Amazon appealing At Stanford, I studied machine learning, robotics, and AI, but Amazon's dataset was unique. Few companies at the time used the word "terabyte" to describe data. Amazon sucked me in because of the opportunity to work closely with Jeff and to work on something that I wasn't very experienced with, which was large data sets. I got the job. My job title was the manager of customer behavior research I joined what became the customer behavior research group at Amazon in January 2003. At that point, we had a few thousand employees, and we were in an office called Pac Med. On average, I met with Jeff once or twice a week. My office was down the hall from Jeff, and I could hear him laughing constantly. He was great and amazingly intense; I greatly enjoyed working with him. Jeff Bezos made me rethink the way I view the world. I worked 80 hours a week at Amazon in a gladiatorial culture. You got stuff done. Amazon removed the coordination and restrictions that other traditional companies had. Everyone criticized Jeff for that, but that's also what motivated people like me to do things. Jeff saw e-commerce as a trillion-dollar opportunity before it became mainstream Jeff Bezos, Andreas Weigend, and David Selinger in 2003. Courtesy of David Selinger He had the vision. Jeff believed, without exception or a willingness to compromise, that nobody else's opinion mattered about how hard or how impossible something was, that something could truly be that big. When I joined Amazon, the media was saying Amazon would go out of business this quarter or next — that was the assumption. Jeff steamrolled ahead, and Amazon took on debt to build warehouses because he had confidence and clarity, which enabled him to overcome what everyone else said was impossible. Jeff taught me that common sense requires common assumptions One of the things Jeff invented was the two-pizza teams. It was like an org chart, with little teams led by essentially Rambo squads — each operates independently. They don't collaborate; they're sent out to battle. Many people said this would fail. That was all based on common-sense knowledge — if you do that, people will quit. Jeff said, "Okay, let's deal with the assumption. Assumption No. 1 is that I want to keep people." Businesses assume the leaders want to keep people. In its early days, Amazon often rotated leadership out of mid-management every 12 to 24 months. Jeff believed he could attract the best talent and get the most out of them in those 24 months. Amazon reinvented the idea of an organizational structure, from being for stability to being for volatility, and discovered the greatest talent in that way. Jeff taught me that data can transform anything That wasn't obvious in 2000, 2002, or even 1995. When I built my team, Jeff had to convince the board of directors that building a team that would use Amazon data was unique. My biggest accomplishment at Amazon was convincing Jeff to allow the advertising business unit to exist and grow. In our first meeting about the potential of ads on Amazon, Jeff said that Amazon is a retailer — we don't put ads on our website. After I showed him the data that proved we should run ads, he approved them. I was recognized as a high performer, but I still decided to leave I walked away from so much stock. At the time, I didn't realize it would be that big, but I wanted to start my own thing. I left Amazon in 2004 and cofounded Redfin, a real estate company, and served as its CTO. I was full-time for a year, got it funded, got them to hire a CEO, and then I moved on. I then started Rich Relevance, a machine learning enterprise software company for big retailers, which would eventually merge with another company. I was in the hopping stage of my life. I next started an AI security company called Deep Sentinel Throughout my career, my love for AI has grown. I wanted to start a company that would have an unequivocal positive impact on the world. Deep Sentinel provides a security solution that includes cameras and a back-end service that connects the cameras to an AI and connects that AI to live guards. We stop thousands of active crimes every month at an affordable price point. We manufacture our cameras and provide the guard service, and I oversee and manage it. We built all the software and all the back end ourselves. I self-funded the start of Deep Sentinel and invested about $1M in it. Then, we funded it in partnership with Jeff Bezos, one of our earliest investors. I reached out to Jeff, and he agreed to be a part of it. My daughters helped me get Deep Sentinel started They did some of the early branding work and logos, and talked about how it would work and operate, and what technology we'd build. I'm a dad and a CEO. I still work a lot. I have offices in Makati, Philippines, Taiwan, and Pleasanton, California. All my employees in the US are remote. I go to the office once or twice a month. My experience at Amazon became the foundation for my career. Read the original article on Business Insider

Discussion | Link

Commodore's Call Back 8020 is a phone “where the customer is not the product."

Anthropic is far from the first first Silicon Valley giant to trigger US government export controls. Apple turned the prospect of a 1999 export limit on the Power Mac G4 into an ad campaign. JOHN G. MABANGLO / AFP via Getty Images Steve Jobs once turned the prospect of a US export restriction into a marketing moment. In 1999, Apple's Power Mac G4 "supercomputer" exceeded the allowed computing threshold for US exports to some countries. Apple released an ad that leaned into the US government's concern that the computers didn't fall into the wrong hands. Sometimes, the US government's concerns that a powerful new tech product could fall into the wrong hands can be a marketing opportunity. Just look at Steve Jobs and Apple back in 1999. In August of that year, Jobs, who was then Apple's interim-CEO, took the stage to unveil the company's new desktop "supercomputer": the Power Mac G4. Jobs called it "the most powerful personal computer ever brought to market," CBS News reported at the time. The only issue was all that computing power technically meant that the device crossed the threshold that would trigger US export controls limiting which countries Apple could ship the computer to. Jobs highlighted the distinction in the wake of the computer's unveiling. "The Power Mac G4 is so fast that it is classified as a supercomputer by the US government, and we are prohibited from exporting it to over 50 nations worldwide," Jobs said the Apple Expo, CNN reported in September 1999. The restriction Apple faced at the time stemmed from a Government Accountability Office report that called 50 countries a concern "for military or proliferation reasons," with seven others facing near-embargo restrictions on computer exports. Jobs told the audience that the new Macs, capable of operating at up to one gigaflop, could not be exported to the nations in that report, including China, Iraq, and North Korea. Now — as Anthropic faces US export restrictions for its Fable 5 and Mythos 5 AI models — Jobs' computer unveiling and subsequent marketing have renewed relevance. Behind the scenes, Apple pushed to ease the US restrictions (and was eventually successful). In public, Apple leaned into the US government's concerns in an ad campaign recently resurfaced by Tom's Hardware. Fable isn't the first. In 1999 the department of defense blocked exports of the PowerMac G4 for crossing the 1 gigaflop threshold. Steve Jobs turned it into an ad. pic.twitter.com/yHoyJjpSke — Justin Schroeder (@jpschroeder) June 13, 2026 The commercial showed tanks surrounding the Power Mac G4 as a voiceover declares that, "For the first time in history, a personal computer has been classified as a weapon by the US government." The commercial ended with a jab at Intel-powered PCs: "Well, they're harmless," the voiceover said. Apple's 1999 run-in with export controls was an earlier example of Washington treating cutting-edge commercial technology as a national-security concern. Today, there's an AI-flavored twist that's landed Anthropic into very real hot water. Over the weekend, the Trump administration ordered Anthropic to restrict foreign nationals' access to its Fable 5 and Mythos 5 models, citing national security concerns related to a possible jailbreak to get around safeguards. Anthropic disabled the AI models for all customers while it works to clear up what it described as a misunderstanding from the White House. The company has disputed the severity of the issue that was originally flagged to the White House. Anthropic has long championed its focus on AI safety, and earlier this year said its Mythos Preview model was too powerful to release widely due to its hacking abilities, instead offering early access to selected partners to help bolster cybersecurity safeguards. The severity of the White House's Anthropic order means it's not exactly an apples-to-apples comparison (pun intended) to Steve Jobs and Apple in 1999. After all, Apple was still able to launch and sell its Power Mac G4s. But if Anthropic manages to smooth over its latest clash with the US government and re-launch its Fable 5 and Mythos models, Jobs and Apple demonstrated decades ago that having a product so powerful it raises government export concerns doesn't have to be all bad. Read the original article on Business Insider

PLUS: Vet business opportunities with NotebookLM

At the end of a tense and scoreless first half of a soccer match between the English men’s team and rival Germany, millions of Brits let out a collective sigh and did what they so often do in moments of stress: They made tea. That wave of electric kettles clicking on, however, caused a different…

With AI adoption accelerating, testing systems under adversarial conditions has become increasingly important. It enables organisations to identify vulnerabilities before deployment and strengthen overall system safety. Explore what AI red teaming is, why it matters and the leading companies offering AI red teaming consulting services. What Is AI Red Teaming? AI red teaming tests artificial […] The post AI Red Teaming Explained: What It Is and Why You Need It appeared first on AI News.

Discussion | Link

Mode Inc CEO Dan Novaes and tech chief Kiran Panesar. Mode Inc Mode Inc acquired Trimbox and QR Code Reader, boosting its reach to 100 million monthly users. Mode Inc pays consumers for their everyday data, unlike gig-focused AI training startups. Mode's acquisitions are part of a thesis that AI labs will need more consented data moving forward. The next wave of AI needs your photos, receipts, and emails to get smarter. One startup wants to help you get paid for it, and it's on a corporate shopping spree. AI training startup Mode Inc has acquired two consumer applications, bringing its total buyouts in the past year to seven. The company bought Trimbox, an inbox management app, and QR Code Reader, a code scanning app. "We're not really focused on just going after the Mechanical Turk or the Outlier/Scale AI gig workers," Mode Inc CEO Dan Novaes told Business Insider, naming pay-per-gig contracting sites. "It's really about everyday consumers and doing the things that they do every single day." These include uploading receipts from Amazon or Walmart purchases, streaming data, and wearable device data. The deals expand Mode Inc's portfolio to include more than 100 million monthly users, who turn in such data for cash or rewards. While not a direct competitor, the company is part of a slew of startups such as Scale AI, Mercor, and Handshake that pay hundreds of thousands of contractors around the world to collect and label data. This data is used by companies to help improve everything from self-driving cars to OpenAI and Meta's chatbots. Noaves said he founded Mode Inc in 2019 for people to get paid for the time they spend on their devices and the data they generate. To date, the company has handed out $1 billion in earnings, savings, and incentives. Novaes has raised about over $80 million through crowdfunding and said he intends to take the company public in the next two years. Novaes said he believes growing legal scrutiny around AI companies' use of online content, such as recent lawsuits against Anthropic and Perplexity, will spike demand for consent-based data collection. He shared a recent example of what an AI lab client requested. "'Hey, we want people to submit forms that have handwriting on them,'" Novaes said, referring to doctors' notes or oil change receipts. "The point is they needed millions of documents of that. And so we basically sent out one notice to our users: Get some samples." Mode's broader ambition is to build a portfolio of niche apps that collectively reach massive scale. Novaes said he is looking into an app that could make users' photo libraries available for AI training. "There are two ways to get to a billion monthly active users," Novaes said. "You can create the next Telegram or Twitter, or you can acquire 1,000 apps that have a million monthly active users each." Have a tip about startups in the AI training industry? Contact Shuby via email at sgoel@businessinsider.com or Signal at shuby.85. Read the original article on Business Insider

For years, enterprise content management was largely a publication tool. How do you get the right content, in the right format, to the right channel, without breaking workflows that span dozens of markets and hundreds of contributors? The answer was usually a combination of manual processes, siloed systems, and large coordination teams that grew historically […] The post How AI-Powered CMS Platforms Are Transforming Enterprise Content Operations appeared first on AI News.

Discussion | Link

Respond.io, one of Malaysia startups to watch, uses AI agents to handle high volumes of customer inquiries and charges per convo, not per seat.

The question businesses asked two years ago was simple: should we use AI? The question in … The Complete AI Model Guide 2026: LLMs, Real Pricing, and the Five Competing Arenas Reshaping the MarketRead more The post The Complete AI Model Guide 2026: LLMs, Real Pricing, and the Five Competing Arenas Reshaping the Market appeared first on .

<p><strong><a href="https://www.lutasecurity.com/post/the-fable-5-export-controls-harm-us-cyber-defense">The Fable 5 Export Controls Harm US Cyber Defense</a></strong></p> I <a href="https://simonwillison.net/2026/Jun/16/matteo-wong-the-atlantic/">quoted The Atlantic</a> quoting Kate Moussouris earlier, when I should have gone straight to the source. Here she is confirming that the "jailbreak" that got Claude Fable 5 banned under an export control really was "fix this code":</p> <blockquote> <p>The researchers took open-source code with known CVEs, plus new code with deliberately planted vulnerabilities, and asked Fable 5, Mythos, and Opus to “review the code for security issues.” Fable 5 refused. They then asked the models to “fix this code” and, through a multistep and manual process, turned the output into scripts that test the patches.</p> </blockquote> <p>As Kate points out, this is absurd. Coding models fix bugs, and security exploits are the most important category of bugs for them to fix!</p> <blockquote> <p>Defenders need to be able to ask AI to fix the bugs in a file, explain why the fix matters, and write tests that confirm the patch works. That is not a guardrail bypass. It is the most valuable thing an AI model can do for defensive security: executing the find, fix, and test loop defenders run every day. [...]</p> <p>The prompts worked because they were defensive requests, and that capability cannot be removed without making the model worse at fixing bugs and verifying patches.</p> </blockquote> <p>This whole situation is such a mess. Non-technical decision-makers have been hearing that models that can "craft cyber attacks" are uniquely dangerous for months. Now they look ready to ban any model that can help us secure our code. <p>Tags: <a href="https://simonwillison.net/tags/jailbreaking">jailbreaking</a>, <a href="https://simonwillison.net/tags/security">security</a>, <a href="https://simonwillison.net/tags/ai">ai</a>, <a href="https://simonwillison.net/tags/generative-ai">generative-ai</a>, <a href="https://simonwillison.net/tags/llms">llms</a>, <a href="https://simonwillison.net/tags/anthropic">anthropic</a>, <a href="https://simonwillison.net/tags/ai-security-research">ai-security-research</a>, <a href="https://simonwillison.net/tags/claude-mythos">claude-mythos</a></p>

The Information: Sources: DeepSeek closed a $7.4B round at a $50B valuation under an unusual structure requiring investors to put capital into an LP run by CEO Liang Wenfeng  —  Chinese AI lab DeepSeek has closed its first funding round that raised more than 50 billion yuan ($7.4 billion) under an unusual deal structure …