 |
|
|
|||||
Anthropic is undoubtedly having a moment right now. First came Claude Code, an AI-powered coding tool for developers, in early 2025, which quickly gained a cult following among that community. “You spent your holidays with your family? That’s nice I spent my holidays with Claude Code,” recently posted one tech-policy expert. But most people aren’t developers, let alone know their way around a command-line interface. So last week, Anthropic launched Claude Cowork, which the company calls “Claude Code for the rest of your work.” Available now as a research preview on Anthropics $100-per-month Max plan, Cowork is the best example of what “vibe coding“an AI-powered approach where people use natural language prompts to bring their software ideas to lifecan do. Designed for non-developers, its a desktop app that aims to help regular workers with all kinds of tasks, like organizing files or crunching data. Case in point: Anthropic’s new working agent was largely built by Claude itself, in just a week and a half. The memes write themselves. “Claude, here is a picture of my bank account. claude, make that number go up to $1 billion. make no mistakes,” one X user prompted. “Claude here is my life. all of it. down to the last detail. make me happy. beautiful. successful. make no mistakes,” another posted. “Claude, here are my notes where I keep all of my passwords. here are my bank account details and phone number for 2fa. run my life and make money, wrote another. Make no mistakes. While Claude might not be able to satisfy those demands (yet), AI is undoubtedly turning the workforce on its head. Research shows that 85% of employees globally are saving one to seven hours a week with AI. Yet, Anthropic CEO Dario Amodei warned that we could be “sleepwalking into a white-collar bloodbath,” with AI wiping out huge swathes of entry-level jobs in just one update. Software engineers in particular, have found themselves directly in the impact zone. Yea Im a full-stack engineer, one X user posted. Their stack: Claude, Terminal, and Cursor. But they arent alone. “Bankers, lawyers, and consultants looking at everyone else joining them in the unemployment line after the Claude Cowork release,” quipped another. In fact, no one is marked safe. “I’m assembling a team,” wrote one X user alongside an image of a company leadership team with Claude in every C-suite role. A follow-up post read “just got kicked out of my own company.” “Got told I was ‘slowing everyone down.'” Despite the discourse, data currently shows that theres little evidence for actual AI-caused displacement in the job market. For now, well at least have the popcorn ready for the memes that just wont quit.
Category:
E-Commerce
Eighteen months ago, it was plausible that artificial intelligence might take a different path than social media. Back then, AIs development hadnt consolidated under a small number of big tech firms. Nor had it capitalized on consumer attention, surveilling users, and delivering ads. Unfortunately, the AI industry is now taking a page from the social media playbook and has set its sights on monetizing consumer attention. When OpenAI launched its ChatGPT Search feature in late 2024 and its browser, ChatGPT Atlas, in October 2025, it kicked off a race to capture online behavioral data to power advertising. Its part of a yearslong turnabout by OpenAI, whose CEO Sam Altman once called the combination of ads and AI unsettling and now promises that ads can be deployed in AI apps while preserving trust. The rampant speculation among OpenAI users who believe they see paid placements in ChatGPT responses suggests they are not convinced. In 2024, AI search company Perplexity started experimenting with ads in its offerings. A few months after that, Microsoft introduced ads to its Copilot AI. Googles AI Mode for search now increasingly features ads, as does Amazons Rufus chatbot. As a security expert and data scientist, we see these examples as harbingers of a future where AI companies profit from manipulating their users behavior for the benefit of their advertisers and investors. Its also a reminder that time to steer the direction of AI development away from private exploitation and toward public benefit is quickly running out. The functionality of ChatGPT Search and its Atlas browser is not really new. Meta, commercial AI competitor Perplexity, and even ChatGPT itself have had similar AI search features for years, and both Google and Microsoft beat OpenAI to the punch by integrating AI with their browsers. But OpenAIs business positioning signals a shift. We believe the ChatGPT Search and Atlas announcements are worrisome because there is really only one way to make money on search: the advertising model pioneered ruthlessly by Google. Advertising model Ruled a monopolist in U.S. federal court, Google has earned more than US$1.6 trillion in advertising revenue since 2001. You may think of Google as a web search company, or a streaming video company (YouTube), or an email company (Gmail), or a mobile phone company (Android, Pixel), or maybe even an AI company (Gemini). But those products are ancillary to Googles bottom line. The advertising segment typically accounts for 80% to 90% of its total revenue. Everything else is there to collect users data and direct users attention to its advertising revenue stream. After two decades in this monopoly position, Googles search product is much more tuned to the companys needs than those of its users. When Google Search first arrived decades ago, it was revelatory in its ability to instantly find useful information across the still-nascent web. In 2025, its search result pages are dominated by low-quality and often AI-generated content, spam sites that exist solely to drive traffic to Amazon salesa tactic known as affiliate marketingand paid ad placements, which at times are indistinguishable from organic results. Plenty of advertisers and observers seem to think AI-powered advertising is the future of the ad business. Big Techs AI advertising plans are shaking up the industry. Highly persuasive Paid advertising in AI search, and AI models generally, could look very different from traditional web search. It has the potential to influence your thinking, spending patterns, and even personal beliefs in much more subtle ways. Because AI can engage in active dialogue, addressing your specific questions, concerns, and ideas rather than just filtering static content, its potential for influence is much greater. Its like the difference between reading a textbook and having a conversation with its author. Imagine youre conversing with your AI agent about an upcoming vacation. Did it recommend a particular airline or hotel chain because they really are best for you, or does the company get a kickback for every mention? If you ask abou a political issue, does the model bias its answer based on which political party has paid the company a fee, or based on the bias of the models corporate owners? There is mounting evidence that AI models are at least as effective as people at persuading users to do things. A December 2023 meta-analysis of 121 randomized trials reported that AI models are as good as humans at shifting peoples perceptions, attitudes, and behaviors. A more recent meta-analysis of eight studies similarly concluded there was no significant overall difference in persuasive performance between (large language models) and humans. This influence may go well beyond shaping what products you buy or who you vote for. As with the field of search engine optimization, the incentive for humans to perform for AI models might shape the way people write and communicate with each other. How we express ourselves online is likely to be increasingly directed to win the attention of AIs and earn placement in the responses they return to users. A different way forward Much of this is discouraging, but there is much that can be done to change it. First, its important to recognize that todays AI is fundamentally untrustworthy, for the same reasons that search engines and social media platforms are. The problem is not the technology itself; fast ways to find information and communicate with friends and family can be wonderful capabilities. The problem is the priorities of the corporations who own these platforms and for whose benefit they are operated. Recognize that you dont have control over what data is fed to the AI, who it is shared with and how it is used. Its important to keep that in mind when you connect devices and services to AI platforms, ask them questions, or consider buying or doing the things they suggest. There is also a lot that people can demand of governments to restrain harmful corporate uses of AI. In the U.S., Congress could enshrine consumers rights to control their own personal data, as the EU already has. It could also create a data protection enforcement agency, as essentially every other developed nation has. Governments worldwide could invest in Public AImodels built by public agencies offered universally for public benefit and transparently under public oversight. They could also restrict how corporations can collude to exploit people using AI, for example, by barring advertisements for dangerous products such as cigarettes and requiring disclosure of paid endorsements. Every technology company seeks to differentiate itself from competitors, particularly in an era when yesterdays groundbreaking AI quickly becomes a commodity that will run on any kids phone. One differentiator is in building a trustworthy service. It remains to be seen whether companies such as OpenAI and Anthropic can sustain profitable businesses on the back of subscription AI services like the premium editions of ChatGPT, Plus, and Pro, and Claude Pro. If they are going to continue convincing consumers and businesses to pay for these premium services, they will need to build trust. That will require making real commitments to consumers on transparency, privacy, reliability, and security that are followed through consistently and verifiably. And while no one knows what the future business models for AI will be, we can be certain that consumers do not want to be exploited by AI, secretly or otherwise. Bruce Schneier is an adjunct lecturer in public policy at Harvard Kennedy School. Nathan Sanders is an affiliate at the Berkman Klein Center for Internet & Society at Harvard University. This article is republished from The Conversation under a Creative Commons license. Read the original article.
Category:
E-Commerce
On January 20, Netflix is bringing back the popular talent competition Star Search, with a twist: For the first time in its history, Netflix will let its audience decide the outcome of a show with live voting. However, unlike how shows have done this in the past, audiences wont have to send text messages or call a special number to make their votes count. Instead, viewers will vote with their TVs remote control, or right within the Netflix app if they watch the show on their phones. Netflix hopes that this level of simplicity will help to make live programs like Star Search a lot more exciting, and offer its audience a chance to experience shared watercooler moments that tend to be missing from todays world of hyper-personalized streaming. You can influence the outcome [together with] everyone at the same time, says Netflix member product VP Elmar Nubbemeyer. Youre part of the Zeitgeist at that moment. To bring real-time voting to Star Search, Netflix relied on work it previously did for interactive narrative shows. It also snuck voting tests into David Changs Netflix show, and showed focus groups segments from two fake shows it cooked up for testing purposes. The company even built internal tools that will help it to repurpose live voting and polling for other live events and shows in the future. We are planning more of these types of moments, says Netflix product designer Navin Iyengar. Star Search is really the big unveiling of it. From ‘Bandersnatch’ to Star Search When Star Search debuts Tuesday evening, viewers will have two distinct opportunities to make their voices heard. Once a singer or comedian is done with their performance, a graphic will pop up on screen, encouraging each viewer to give it a rating ranging from one to five stars. We knew early on that giving a star rating as an interaction was really important, says Iyengar. Its core to the Star Search IP. Later on, theyll also get the chance to choose their personal champion of the night out of four choices presented next to each other on screen. Each voting graphic will remain on screen for about 60 seconds, and the shows hostAnthony Anderson, best known for the ABC sitcom Blackishwill respond to the incoming vote tally in real time. Chrissy Teigen, Jelly Roll, Sarah Michelle Gellar, and Anthony Anderson [Photo: Matt Sayles/Netflix 2025] Its the first time Netflix has done real-time voting like this, but the company has been experimenting with getting viewers more actively involved for almost a decade. In 2017, the streaming service released its first interactive TV shows, which prompted viewers to choose their own adventure through branched narratives. In one scene of “Bandersnatch,” an interactive episode of the dystopian sci-fi show Black Mirror, the viewer has to decide whether the main character should take his medication by pressing left or right buttons on their remote control, with different choices leading to vastly varying outcomes. Even in those early days, Netflix engineers and designers already thought about ways to bring the same kind of interactivity to live content. We always felt that interactive experiences should go live, because members could actually interact in the moment and impact the story as it’s happening in real time instead of filming all the different potential outcomes, recalls Iyengar. Netflix eventually discontinued branched narrative shows because they did not take off with consumers, but the company kept pursuing the idea to marry live content with interactivity. And when Star Search came along, it quickly became clear that this was a perfect opportunity to bring back some of that interactive tech first built for titles like Bandersnatch, and use it to improve the way audience participation is usually done. Talent shows like American Idol used to rely heavily on phone calls to register votes, but nowadays use a mix of online and text message voting. Shows often allow participants to vote multiple times, leading to massive vote counts, which often dwarf the number of voters. It’s a surprisingly low share of viewers who actually reach out and vote, Iyengar says, adding that industry estimates put that number somewhere between five and ten percent. Voting has always been difficult for these shows, he adds. Netflix employees believed that the tech first built for Bandersnatch, which allowed viewers to send feedback with their TV remote, would already go a long way towards making it easier to participate. But they quickly realized that great tech alone wasnt enough. Keeping voting fair, even for Star Search Netflix began testing prototypes for live voting with focus groups nearly a year ago. To do so, the company repurposed two existing titlesa dating show and a talent competition. We basically made fake shows, Iyengar says. We edited them down to make them feel live and make it feel like your vote was really important. Then, it put select viewers into a lab designed to look like a living room, with a double-sided mirror to observe how they reacted. We put our prototypes on a TV, Iyengar says. We had a TV remote that could control it. People were sitting on a couch, and we would actually just leave them alone. Netflix researchers just told test participants that they get to take a peek at a prototype, without explaining that the show would allow them to vote. People really got it, Iyengar recalls. Almost everyone, without prompting from us, would pick up the remote in these moments, and interact. In fact, people didnt just get itthey got hooked. We found that they got really invested in the stakes of the show, even though it was fake, Iyengar says. They wanted to know what their vote was going to do. Did the person I voted for win? Show me the math for how you actually calculated the vote. People just took the idea of voting and fairness seriously. Netflix built its voting tech to only allow one vote per Netflix profile. But those early tests showed that fairness was as much about the way different options were presented, and that long held beliefs about UI design could introduce perceived biases. One example: Designers like Iyengar like to direct the eye to simplify smart TV interfaces. When you open up the Netflix app on your TV, youll find that one title is always pre-selected, which helps to understand what to do if you want to navigate to the title right next to it, or perhap one in a row below. On TV, you should always have something in focus, he says. Otherwise people don’t understand where the focus state is. When Iyengars team built the interface people will use to award stars to Star Search performers, they initially followed that same principle, and highlighted the third star to direct the eye. Test audiences immediately pushed back. People did not like that we were filling up the stars for them, he says. They were like: why are you voting for me? Voting on tuna sandwiches and sports competitions? In addition to the interface, Netflix also built a dedicated tool called Pollster that allows producers to integrate voting into their shows, then trigger each round in real time. To test Pollster and the backend tech for voting ahead of this weeks Star Search premiere, the streamer snuck a few test votes into Diner Time Live, a live cooking show hosted by celebrity chef David Chang. Diner Time Live is not a competition, so testing star ratings didnt really make sense. David Changs team nonetheless embraced the idea, and let the shows audience rate different kinds of sandwiches. It wasn’t something high stakes, Iyengar says. Audience participation was nonetheless high. I’m really glad we did it, he says. We learned a lot. The streamer is now ready to put those lessons learned to the test with the premiere of Star Search on Tuesday and already has plans to bring it to additional live entertainment formats in the future. We have many other ideas where we could apply this technology, Nubbemeyer says. One of the things were [considering] is polling. Netflix may, for instance, use its voting tech to ask viewers of a sports event to decide who the most impactful player is. All these things could enrich the entertainment experience by making it more participatory, Nubbemeyer says.
Category:
E-Commerce
All news |
||||||||||||||||||
|
||||||||||||||||||