 |
|
|
|||||
Amaras Law, coined by the American scientist and futurist Roy Amara, says humans tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run. If the first half of 2025 is anything to go by, in the AI era, the runs are getting shorter, and the effects of the technology will be larger than weve seen in a generation. In a matter of months, the conversation in companies has accelerated far beyond if AI is a useful productivity tool, to where and when it can be applied. Across industries and geographies, executives are acknowledging that AI is a general-purpose business solution, not just a technical one. Despite widespread workplace adoption, the focus on cybersecurity has not kept pace. In the rush to adopt AI systems, applications and agents, companies are failing to consider that rapid deployment of these new technologies could lead to data breaches and other security risks. That matters because AI models are not only getting more powerful but also more useful for enterprises. More enterprises are using AI agents As of early June, OpenAIs base of paying business users reached 3 million, up from 2 million in February. In a move for that market, ChatGPT can now connect to popular business apps such as Google Drive, Dropbox, and Sharepoint, allowing workers to quickly access answers that are locked in dispersed documents and spreadsheets. Confusion, and even fear, about AI agents has given way to exploration and adoption. Among US-based organizations with annual revenues of $1 billion or more, 65% were piloting AI agents in the first quarter of this year, up from 37% in the space of a single quarter. Microsofts Azure AI Foundry, its platform for building AI agents, processed 100 trillion tokens in the first three months of 2025 (with one token representing the smallest unit of text that an AI model processes)a five-fold increase year-on-year. At the same time, the cost per token more than halved, spurring higher use and creating virtuous cycles of innovation. As John Chambers, the former CEO of Cisco, says, AI is this generations internet revolution but at five times the speed, with three times the outcome. Beyond the hype that haunts the sector, there are signs of enterprise AI adoption everywhere. In his latest letter to shareholders Alex Karp, CEO of Palantir Technologies, describes a ravenous whirlwind of adoption of AI. IBM, which has rolled out its AI strategy to 270,000 employees, reports that AI already handles 94% of routine human resources tasks. At Shopify, the e-commerce group, AI usage is now a baseline expectation, CEO Tobias Lütke said in an employee memo. The workplace automation company Zapier, which took steps to embed AI across its workforce, says that 89% of employees actively use AI in their daily work. The list goes onand its not just technology companies. JP Morgan, the worlds largest bank, has rolled out GenAI tools to 200,000 staff members, and says employees have each gained one-to-two hours of productivity each week. AI acquisitions are plentiful The shift from novel to mass-market tech is reflected in the business strategies of the main AI model makers, which are reimagining themselves as application companies. In the space of two weeks, OpenAI, the ChatGPT parent, appointed a CEO of Applications and then acquired IO, the AI device startup founded by former Apple designer Jony Ive, for $6.5 billion. Meta, perceived to be behind in the AI race, has invested $14.3 billion in Scale AI, which provides data and evaluation services to develop applications for AI. Meanwhile, Apple is reported to have had internal talks about buying Perplexity AI, a two-and-a-half year-old AI model maker. AI app security is rarely discussed Companies are naturally focused on the potential and performance of AI systems, but it’s striking how rarely security is part of the story. The reality is that the speed of deployment of AI apps and agents is leaving companies at risk for breaches, data loss, and brand impact. For example, an AI system or agent that has access to employee HR data or a banks internal systems leaves a company open to possible cyberattacks by bad actors. In business-critical applications, risks emerge at every stage of the development cycle, from choosing which AI model to use and what systems to give it access to, right through to deployment and daily use. In our work on testing the security of AI models with simulated attacksknown as red-teamingand creating the CalypsoAI Model Security Leaderboards, we have discovered that, despite performance improvements, new or updated AI models are often less secure than existing ones. At the same time, existing models can see their security score slip over time. Why? Because the attacks keep progressing and bad actors learn new tricks. More techniques and capabilities of breaking or bypassing AI model securities keep being invented. Simply, the attack techniques are getting better and they’re causing AI models that have only recently launched to become less secure. That means that organizations that begin using an AI system or agent today, but don’t stay up to date with the latest threat intel, will be more vulnerable as attack techniques increase in capability and frequency. As corporate AI systems gain autonomy and access to sensitive data, what is safe today may not be safe tomorrow. The research firm Gartner has forecast that 15% of day-to-day business decisions will be made autonomously by agents by 2028, though that percentage may increase by then. Against that backdrop, virtually all the security protocols and permissions in enterprises are built for human workers, not for AI agents that can roam through company networks and learn on the job. That mismatch opens up vulnerabilities, such as the ossibility of agents accessing sensitive information and sharing it inappropriately. Poorly secured agents will be prime targets for hackers, particularly where they have access to valuable data or functions such as money transfers. The consequences include financial loss and reputational damage. Final thoughts Securing these new systems will be critical to AI adoption and to successful return on investment for the companies involved. A new security paradigm, using the capabilities of agentic AI to secure enterprise AI, is needed to allow innovation to thrive and agents to reach their potential. While the development of AI models and systems so far can reasonably be summarized as better, cheaper, less secure, the final part of that equation must improve significantly as the emerging application-first AI era accelerates. Once that happens, Roy Amara seems certain to be proven right once again. Donnchadh Casey is CEO of CalypsoAI.
Category:
E-Commerce
As Zohran Mamdani declared victory in the New York City Democratic mayoral primary on Tuesday night, one had to wonder: Has anyone checked on the finance bros? On X, the Wall Street meltdown was already well underway. It appears that NYC is electing to commit suicide by Mayor, wrote Jim Bianco, president of Bianco Research. Its officially hot commie summer, added Dan Loeb, CEO of hedge fund Third Point and longtime Cuomo backer. every NYC finance guy on my main feed right now pic.twitter.com/Z7anouAFio— (@EffMktHype) June 25, 2025 Loeb wasnt alone. Billionaires like Michael Bloomberg and Bill Ackman had backed Andrew Cuomo, still seen as the frontrunner even in the races final days. Bill Ackman drafting his thoughts on Mamdani rn, one post joked, alongside an image of an essay-length text being written. Another great part of Mamdanis victory is that it means Michael Bloomberg pretty much lit over $8 million on fire for no reason lol, another X user wrote. Bill Ackman drafting his thoughts on Mamdani rn pic.twitter.com/e7pCMMKKO4— litquidity (@litcapital) June 25, 2025 The finance industrys reaction isnt surprising. A Mamdani win in Novembers general election could bring what Wall Street dreads most: tax hikes and tighter regulations threatening corporate and investment interestsfueling the familiar cry of a wealthy exodus. Wealthy New Yorkers moving to Miami if Zohran wins, one Instagram meme joked. View this post on Instagram A post shared by New Yorkers (@newyorkers) Lets not forget the finance bro who did vote for Mamdani (starter pack includes a Carhartt beanie and a copy of The Communist Manifesto). NYC girls with trust funds were calling him Zaddy Zohran and you thought he was going to lose? one user posted on X. pic.twitter.com/0mZY1Kfw2s— Overheard on Wall Street (@OHWallStreet) June 24, 2025 One post, acknowledging defeat, featured an AI-generated image of fleece-clad finance bros scanning groceries at a city-run market: Well boys, onto the new bullpen. Well boys, onto the new bullpen. pic.twitter.com/GUxzYaYZw6— Overheard on Wall Street (@OHWallStreet) June 25, 2025 Another post on X perfectly summed up the general mood: Investment bankers reacting to NYC nominating a socialist for Mayor. The accompanying caption? wed all be fine with a lot less, right? Investment bankers reacting to NYC nominating a socialist for Mayor pic.twitter.com/4SY66q4qTn— litquidity (@litcapital) June 25, 2025
Category:
E-Commerce
In my conversations with business leaders around the world, I consistently hear the same phrase to describe what they want to achieve for their workforce: AI fluency. I often tell them that to achieve AI fluency, we need to treat it as a foreign language. Like learning a new language, becoming AI fluent requires dedication, immersion, and practice. Fluency transforms how we think and communicate. Becoming fluent requires us to overcome the fear of making mistakes or incurring risks. Yet theres one crucial difference between achieving fluency in AI versus a new language: When learning a new language, we step into an established culture. With AI, were learning the culture while simultaneously creating it. The big question: How can organizations build these cultures and become laboratories of AI fluency? Here are three ways to foster AI fluency. 1. Create an immersive environment Whether were learning Spanish, Mandarin, or any of the other 7,000 languages in the world, immersion is an essential step to fluency. Living where the language is spoken forces you to adapt, to think differently, and to develop new neural pathways. AI requires the same commitment. Organizations are uniquely positioned to create these immersive environments where employees interact with AI tools daily, not as occasional novelties but as essential components of their workflow. From Udemys work with thousands of organizations around the world, helping to create these environments, weve found that organizations succeed when they integrate AI across departments, from marketing teams using generative AI for content creation to HR departments employing AI-powered skills assessments. Immersive environments are built when employees understand they need to become fluent to reach their goals. That means the most successful AI adoption happens when tools directly address employees pain points. Just as language learners progress faster when they need the right words to order food or navigate transportation, employees embrace AI more readily when it solves real problems they face. Organizations seeking AI fluency must balance structure with exploration. Consider how language learning works: Structured lessons provide grammar and vocabulary, but real learning happens through conversation and experimentation. Similarly, building organizational AI fluency requires a few basic building blocks: Upskilling on foundational AI capabilities and limitations, like learning the rules of grammar. Creating a sandbox-style environment where people can experiment without fear of consequences. Developing communities of practice where people can find social support to troubleshoot, ask questions, celebrate successes, and motivate each other to keep experimenting. Establishing guidelines for when to rely on human judgment versus AI, how to evaluate AI outputs, and how to maintain human connection in AI-mediated interactions. 2. Overcome fluency barriers The barriers to AI fluency mirror those of language learning. Fear of embarrassment prevents many language learners from practicing conversation, just as fear of looking incompetent may prevent employees from experimenting with AI. Imposter syndromethe feeling that everyone else knows more than you doimpacts both AI and language fluency. The solution is creating psychologically safe environments where questions are welcomed, and mistakes are treated as learning opportunities. Leaders like Salesforce CEO Marc Benioff model this by encouraging employees to approach new challenges with a beginners mind, getting curious instead of expecting immediate mastery and understanding. Whats more, both language learners and AI adopters often experience an uncanny valley stage where they know enough to recognize their limitations but not enough to feel confident. Supporting people through this phase is critical. This is where many abandon the journey if theyre not properly encouraged. In this case, encouragement can come not only from leaders, but from the environments leaders create such as building supportive communities of practice where learners can share their struggles with gaining fluency. This normalizes the experience, while reminding them that this uncomfortable stage is not just common but also a sign of meaningful growth. 3. Create culture while learning This is where the language metaphor ends. While becoming AI fluent, were simultaneously students and architects of the culture. This dual role presents unprecedented responsibility and opportunity. Leaders must consciously shape how AI integrates into the organizational culture by establishing rules and norms that preserve human creativity and connection while leveraging AIs capabilities. This means modeling thoughtful AI usage, celebrating innovative applications, and continuously reinforcing that AI serves human objectives, not the reverse. The organizations that thrive will be those that build immersive environments where employees can become AI fluent and build cultures where technology amplifies uniquely human capabilities. In a workplace where managers offload administrative or basic creative tasks to AI agents, employees would gain hours back in their day. This would allow them to spend more time coaching their teams, helping them solve problems, identify opportunities for growth, and learn the best ways to motivate them during times of change and upheaval. The journey to this future begins with recognizing that AI, like any language, isnt just a skill to acquire but a new way of thinking. Hugo Sarrazin is the CEO of Udemy.
Category:
E-Commerce
Donald Trump is no stranger to selling unusual merch. The president has been known to slap his likeness on Bibles, guitars, and gold-plated phones if itll bring in a bit of cash. Just last year, the Trump Organizations licensing deals netted Trump tens of millions of dollars in profits. But today, Trump has outdone himself with what has to be his strangest piece of merch ever: a T-shirt featuring his own mug shot captioned with the word Daddy (in call caps). The official Trump Daddy shirt, featured in an email newsletter sent to subscribers on June 26 from the Trump National Committee JFC, appears to be a limited-time offering separate from the official Trump store. Its being sold through its own webpage with the notice These are going to sell FAST. HURRY, claim your shirt today, as well as the somewhat more unsettling all-caps call to action CLAIM DADDY SHIRT. [Screenshot: Winred] The merch-ification of Trumps mug shot is a fairly tired tack at this point, given that the president has already sold pieces of the suit worn in the image, reprinted it on rally posters, and used it as inspiration for his official White House portrait. Earlier this month, Cara Finnegan, author of the book Photographic Presidents: Making History From Daguerreotype to Digital, told Fast Company that the mug shot has been used so often by Trumps administration that it arguably is his presidential portrait. However, the use of the moniker Daddy is altogether new. It seems to be a nod to a comment made June 25 by Mark Rutte, secretary-general of NATO, who appeared to give Trump the title when he said, “Daddy has to sometimes use strong language. The media (and, now, it seems, Trumps team) took the comment as a reference to a viral moment earlier in the week when Trump casually dropped the f-bomb on camera. Rutte has since denied that the term was directed at Trumpbut, unfortunately, that didnt stop the White House from posting an edit of the President set to Ushers sexual innuendo-laden R&B song, “Hey Daddy (Daddys Home). Trump himself appeared to endorse Ritters use to the term daddy at a later press conference. Now, it seems, the Trump National Committee is shortening the life cycle of real news-turned meme-turned merch. Mere hours after the initial stir around the daddy comment, its already a shirt. Its not entirely shocking to see Trumps fundraising committee capitalize on this moment, given that his administration has already thrown convention to the wind by constantly glamorizing his mug shot. Still, this development is markedly more unusualand if this flies as a kind of Presidential merch, we dont want to know what might appear on a Trump shirt next.
Category:
E-Commerce
A first public test of robotaxis by Tesla in Austin, Texas led to multiple traffic problems and driving issues, videos from company-selected riders showed over the first few days. Chief Executive Elon Musk has tied Tesla’s financial future to self-driving technology, and with Tesla sales down, the stakes are high. He said Tesla would roll out the service to other U.S. cities later this year and predicted “millions of Teslas” operating “fully autonomously” by the second half of next year. The Tesla fans invited to the trial were strongly supportive and posted videos of hours of trouble-free driving, but issues drew questions from federal road safety regulators and auto safety experts. Issues included Tesla robotaxis entering the wrong lane, dropping passengers off in the middle of multiple-lane roads or at intersections, sudden braking, speeding and driving over a curb. In one instance, a robotaxi drove into a lane meant for oncoming traffic for about 6 seconds. It had pulled into an intersection in its left-turn lane with its turn blinker on. Then the steering wheel wobbled momentarily, and instead of turning it proceeded straight into the lane meant for oncoming traffic, prompting a honk from a car behind it. In another incident, the car suddenly braked with no obstruction apparent in the video. The passenger jerked forward and their belongings were thrown to the floor. In a third video, taken from another vehicle, a robotaxi abruptly stopped twice in the middle of the road while passing police vehicles with flashing lights. Tesla is conducting the test with human safety monitors in the front passenger seat. A fourth video showed the safety monitor hitting a button to stop the robotaxi when a delivery truck in front of it started backing up. “This is awfully early to have a bunch of videos of erratic and poor driving,” said Philip Koopman, a Carnegie Mellon University computer-engineering professor and autonomous-technology expert. “I was not expecting as many videos of problematic driving on the very first day,” he said. Tesla is testing about 10 to 20 robotaxis, which are standard Model Ys with advanced software, and has been giving rides since Sunday afternoon. Reuters was able to independently verify the locations of at least 11 videos showing issues. Tesla did not respond to a request for comment. A City of Austin spokesperson said officials are aware of the Tesla issues documented on social media and that “when a potential legal or safety concern is brought to our attention, we promptly share it with the company.” The spokesperson added that the police department is “actively collaborating with Tesla” to ensure officers can safely interact with the robotaxis. ‘Caught on camera’ The incidents caught on camera did not involve accidents, and one expert said some reflected a decision to focus on safety. “So far so good. It handled the situations very well and likely better than even good drivers,” Alain Kornhauser, Princeton University professor of operations research and financial engineering, said by email. He added that it would be more dangerous to drive at less than the speed of prevailing traffic, for instance. Tesla’s experiment is unusually public. Other companies faced similar issues: Alphabet’s Waymo and General Motors’ Cruise had their own share of traffic mishaps after showing up on Austin streets. City officials logged dozens of instances over the past two years where residents and authorities reported that robotaxis blocked traffic by stopping in the middle of roads, failed to respond to police directions and could not deal with emergency vehicles and road closures. A serious accident involving a pedestrian in 2023 led Cruise to shut down last year. Waymo is the only robotaxi service in the U.S. to ferry paying customers without a human backup driver or in-car safety monitor. It started offering rides to the general public through Uber in Austin earlier this year. Musk for years has failed to deliver on promises that self-driving Teslas are just around the corner. Tesla rolled out the service for a flat fee of $4.20 to a limited number of handpicked riders. The service is not available to the broader public and the robotaxis operate in a limited area, and avoid difficult intersections and bad weather. Riders were rarely bothered much by driving issues. Farzad Mesbahi, a former Tesla program manager, and his co-passenger hit the “drop off early” option during a ride. The vehicle stopped in an intersection with a stoplight, his video showed. They exit quickly and walk to the sidewalk. “The car should have known to not stop there,” Mesbahi is heard saying after the ride. “Opportunities for improvement,” the co-passenger says. That is an example “most companies would not be comfortable with,” said Kara Kockelman, a professor of transportation engineering at the University of Texas at Austin, adding that she was surprised by the traffic mistakes. “Dropping off people in the middle of a six-lane road or edge of a busy intersection when the traffic is going in the opposite direction is pretty dangerous. They definitely did not want to do this or be caught on camera,” she said. Abhirup Roy, Rachael Levy, and Chris Kirkham, Reuters Additional reporting by Inaki Malvido, Fernando Robles, and Richa Singh.
Category:
E-Commerce
Sites : [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] next »