Latest news with #SiliconValley
Yahoo
31 minutes ago
- Business
- Yahoo
AI Researcher Declines $1 Billion Offer From Mark Zuckerberg
Meta CEO Mark Zuckerberg is absolutely tripping over himself to attract top AI talent to fill the ranks of his recently announced Superintelligence Labs — and he's willing to go to obscene lengths to make it happen. As Wired reports, more than a dozen people at Thinking Machines Lab, an AI startup founded by former OpenAI executive Mira Murati, were approached by Meta with ludicrous job offers. One individual was even offered more than $1 billion, paid out over several years, a source told Wired — a staggering amount of money, needless to say. But despite the mind-shattering amount of money on the table, not a single person at TML has taken the offer, hinting at some glaring brand issues Zuckerberg could be dealing with. Ouch! Meta communications director Andy Stone disputed the exact figures to Wired, but confirmed offers were made. The reporting suggests we've truly hit peak AI hype, as hundreds of billions of dollars are being poured into the tech despite little in the way of ROI in sight. The news comes as an increasingly desperate Zuckerberg has been offering enormous sign-on bonuses to flesh out his Superintelligence Labs, as part of a broader bid to stay relevant in the ongoing race currently being led by Meta's competitors, including OpenAI and Google. Earlier this year, Zuckerberg poached Ruoming Pang, the former head of Apple's fumbling "Apple Intelligence" software, as well as 28-year-old Scale AI cofounder Alexandr Wang, to lead the lab. "We've been following your work on advancing technology and the benefits of AI for everyone over the years," reads a message Zuckerberg previously sent to a potential recruit, as quoted by Wired. "We're making some important investments across research, products and our infrastructure in order to build the most valuable AI products and services for people." Meta has struggled to keep up with the competition, stumbling with the rollout of its latest Llama 4 large language model. Earlier this year, the company became embroiled in controversy after being accused of fudging the AI model's benchmark results. That general aura could have to do with why the TML staffers didn't take Meta up on its more-than-generous offers. For one, sources told Wired that they're not excited about Wang's leadership style, raising concerns over his "relative lack of experience." The Murati-led startup also raised the largest funding round in history, ballooning to a valuation of $12 billion in just a single year. In other words, they have plenty of money to throw at talent retention. As the AI hype continues to balloon to epic proportions, Zuckerberg is spending billions of dollars on AI talent to realize his hazy goal of creating what he calls "personal superintelligence." It's an enormous bet, as investors and economists are becoming increasingly wary of a dot-com-style bubble in the AI space, where the tech is exciting-yet-imperfect, and a sustainable business model remains elusive. If anything, the latest news suggests AI researchers are in the driver's seat — and Zuckerberg's Meta is clearly not making the list as they continue to collect cushy job offers that could make them wildly wealthy overnight. More on Meta: Mark Zuckerberg Looks Like He's Been Taken Hostage as He Explains Plan for Deploying AI Superintelligence Errore nel recupero dei dati Effettua l'accesso per consultare il tuo portafoglio Errore nel recupero dei dati Errore nel recupero dei dati Errore nel recupero dei dati Errore nel recupero dei dati
Digital Trends
an hour ago
- Entertainment
- Digital Trends
Casetify cover puts the charm of centuries-old art on the AirPods Max
Apple's first over-the-ear headphones have stirred quite a polarizing debate over minimalist looks and a familiar brutalist industrial design language. You can't quite call them artsy, or a piece of art, for that matter. But if you've ever wished to give them such a makeover, casemaker Casetify went straight for the summit with its headphone cover. The company calls it 'Girl with a Pearl Earring Collectible Headphone Cover,' and has listed it at a sweet $199, enough to buy a pair of half-decent noise-cancelling headphones. The ambitions are so high that the accessory maker won't entertain any promotional coupons or discounts for this AirPods Max cover. It seems that sky-high premium for a fabric cover has little to do with any uber exotic fabric material or technical wizardry, and more about the inspiration behind it. To be specific, the cover is modeled after the famous 'Girl with a Pearl Earring' painting by famed Dutch artist Johannes Vermeer. Recommended Videos The fancy headphones cover borrows the signature blue cover and silky look from the 17th-century painting, and even has pearl earrings dangling from each side. It's quite a sight to behold, but I'm not sure if an average person would like to go out wearing them, unless their name is Rihanna or Mariah Carey, and the venue is a red carpet in Milan. This won't be the first such adventure of its kind for Casetify. The company recently collaborated with The Metropolitan Museum of Art and released cases featuring famous paintings from the collection, alongside the Jacques-Louis David 'The Death of Socrates' Limited Canvas Edition Premium Set Box. The company also launched protective gear in partnership with the National Museum of Korea, and hear me out, the Museum of Ice Cream. Interestingly, this won't be the first time that the brand has made cool headphone gear. The limited edition DRAGON BALL Z SCOUTER Collectible Headphone Case was quite a sight to behold. Take a look: The latest AirPods Max cover from Casetify is part of a collaboration with The Mauritshuis Museum in The Hague. You can check out the entire collection featuring protective cases for iPhones, phone charms, and watch bands here. If art is not your thing, Casetify has also partnered with entertainment franchises such as Dragon Ball-Z, One Piece, Smurfs, Peanuts, Squid Game, Harry Potter, BMW, NBA, and more. Of course, K-pop, too. You can check out all the Casetify Co-Lab products on the brand's hub.
Yahoo
2 hours ago
- Business
- Yahoo
Top analyst sends Apple CEO blunt message about its future
Top analyst sends Apple CEO blunt message about its future originally appeared on TheStreet. Apple's reputation for creating groundbreaking consumer electronics and software under former founder and CEO Steve Jobs is being tested this year. CEO Tim Cook is under intense scrutiny to make a splash in artificial intelligence, a massive market opportunity already being captured by deep-pocketed rivals, including OpenAI's ChatGPT, Google's Gemini, and Microsoft's Copilot. 💵💰💰💵 So far, Apple's foray into AI with Apple Intelligence has been criticized for being too little, too late. Apple watchers are increasingly urging Cook to take action and boldly move in the space, reminiscent of Jobs' iconic "one more thing" reveals at Apple conferences. Veteran Wall Street analyst Dan Ives is the latest to weigh in on Apple's AI problem. The long-time Apple analyst delivered a blunt message on what Apple should do next to address its AI ambitions. Apple's future depends on thinking big on AI The flurry of AI chatbot activity is already upending how consumers and businesses consume information. Rather than digging through endless search results to find answers or information, people are increasingly turning to large language models like ChatGPT, Gemini, and Copilot to get deeper and robust answers more also using it to create in ways unthinkable two years ago. People use it to research deeply, produce and consume entertainment, and help organize projects. Businesses are building custom AI agents to boost worker efficiency, improve sales, and eliminate customer service bottlenecks. AI's applications are seemingly endless, leading many to believe AI chatbots and agents will be the most disruptive technological advance since the internet. As a result, huge amounts of money are at stake. The IDC FutureScape report predicts global spending on technology to support AI will reach $337 billion in 2025 and grow to at least $749 billion by 2028 as companies shift from building AI infrastructure and training generative and agentic AI to inference or using AI apps in the real world. Capturing its fair share of that growing pie is going to require Apple making some bold moves, according to Ives, who has been tracking technology companies for decades and is currently Managing Director and Senior Equity Research Analyst at Wedbush Securities. Following Apple's $100 billion additional commitment to U.S. manufacturing, Ives offered a blunt three-prong approach Apple should use to accelerate its AI roadmap. "In our view: Three Moves Cook & Cupertino Need to Make To Avoid a 'Blackberry Moment' in AI. Address the elephant in the room…🎯🏆👇🍎," wrote Ives on X, formerly Twitter. His three ideas include: "1. Do the Perplexity acquisition before it's too late. 2. Bring in outside talent to the management team. 3. Do the Google Gemini partnership." Apple has reportedly considered using its $36 billion cash war chest to acquire Perplexity, an existing AI player. Adding more top-notch AI talent could also help it leapfrog rivals, as other big tech companies are already doing. Finally, partnering with Google Gemini for a more robust AI experience on iPhones, tablets, and Macs could help it close the AI gap for users. Apple's next AI moves are critical (and likely costly) Apple doesn't have a shortage of financial firepower to follow Ives' roadmap. It boasts some of the most loyal consumers on the planet, and racked up an eye-popping $94 billion in sales in the June quarter, resulting in $23 billion in net income. Still, following advice won't come Cook wants to accelerate Apple's AI development, acquiring Perplexity could do the trick. Perplexity markets an AI chatbot similar to Google Gemini and OpenAI's ChatGPT. Its in-house model, Sonar, is built on Meta Platforms' Llama open source code. Perplexity has also leveraged DeepSeek's R1, and Perplexity Pro subscribers can choose between other large language models, including OpenAI, Gemini, Anthropic's Claude, and Elon Musk's Grok. Perplexity's user-first approach of matching the right Chatbot and real-time insight to users' needs nicely matches Apple's historical mission. It would provide an out-of-the-box solution for Apple to monetize AI. Perplexity, launched in 2022, has over 22 million monthly active users. According to About Chromebooks data, its annualized revenue could reach $150 million this summer, up from $63 million last year. More AI Stocks: Jim Cramer drops jaw-dropping price target on Palantir stock post-earnings Morgan Stanley slaps eye-popping price target on Nvidia stock Apple CEO drops bombshell about its future Perplexity isn't publicly traded and, with an AI search market share of about 6%, is reportedly valued at $18 billion. Apple would likely have to pay much more than that, but Cook may be willing to consider it. 'We've acquired around seven companies this year,' said Cook on Apple's recent earnings conference call. 'We're very open to M&A that accelerates our road map… We are not stuck on a certain size company." Poaching away AI talent would be costly, too, given recent deals. An AI talent war has broken out this year. For instance, Anthropic, backed by Amazon, is aggressively hiring and working to retain workers. OpenAI's Sam Altman has accused Meta of trying to poach its workers by offering $100 million bonuses. This summer, Meta Platforms invested $14.8 billion in Scale AI to acquire a 49% nonvoting stake and lock up Scale AI CEO Alexandr Wang to lead its new superintelligence project. Inking a deal with Gemini looks like it could happen soon. The Verge reported earlier this year that during Google's search monopoly trial, Alphabet CEO Sundar Pichai confirmed a meeting with Cook and suggested a deal to get Gemini on Apple devices could come this year. It remains to be seen what Apple's AI plans ultimately are, but one thing seems clear: It needs to act soon if it wants to avoid falling too far analyst sends Apple CEO blunt message about its future first appeared on TheStreet on Aug 8, 2025 This story was originally reported by TheStreet on Aug 8, 2025, where it first appeared. Sign in to access your portfolio
Reuters
3 hours ago
- Business
- Reuters
US retailer Best Buy weighs boosting India headcount, Subramanian says (Aug 8)
CHENNAI, Aug 8 (Reuters) - U.S. retailer Best Buy (BBY.N), opens new tab is weighing the expansion of its India headcount to primarily add more digital and technology roles, Nithya Subramanian, senior director, data & AI COE, told Reuters. Best Buy employs around 350 people at its global capability centre in Bengaluru, better known as "India's Silicon Valley". That headcount could grow to approximately 500 over the next few months, Subramanian said on the sidelines of an event in Chennai. Many global companies have been setting up offices or boosting their presence in India to tap its growing talent pool. GCCs have been evolving into high-value innovation hubs from low-cost back offices in recent years. They now support their parent firms in vital functions including operations, finance, and research and development. "We will be hiring across the functions," Subramanian said. Best Buy, known for selling electronic items such as laptops, kitchen appliances and cameras, is looking to hire AI engineers, software engineers and product managers in India, according to its LinkedIn page. "Even if you look at the global strength, I think we are growing leaps and bounds in India," Subramanian said, noting that the Bengaluru office is Best Buy's largest tech hub. Best Buy operates more than 1,000 stores in the United States and Canada, where it employs over 85,000 people. It does not have retail operations in India. Another U.S. retailer, Costco Wholesale (COST.O), opens new tab, is gearing up to open its first India GCC, sources told Reuters last month. India's GCC market is expected to reach between $99 billion and $105 billion by 2030, from $64.6 billion in fiscal 2024, a report by industry body Nasscom and consulting firm Zinnov said.
Geeky Gadgets
6 hours ago
- Geeky Gadgets
ChatGPT 5 vs Claude Sonnet : AI Coding Skills Compared
What if the future of coding wasn't just about human ingenuity but also about the battle of AI giants? Imagine two innovative AI models—OpenAI's GPT-5 and Anthropic's Claude Sonnet—competing to build a multiplayer tic-tac-toe game. One delivers deliberate, rule-abiding solutions, while the other dazzles with speed and visual finesse. Yet, neither emerges flawless. This isn't just a tech showdown; it's a glimpse into how AI shapes the tools we use to solve problems, design interfaces, and even manage costs. As developers increasingly rely on AI for coding tasks, the stakes have never been higher: which model truly delivers the edge? In this comparison , Convex breaks down the strengths, weaknesses, and surprising quirks of GPT-5 and Claude Sonnet. From token efficiency and pricing to their ability to handle complex tasks like authentication, you'll discover how these models stack up in real-world scenarios. Whether you're a developer seeking precision or speed, or just curious about how AI models differ in their approach to problem-solving, this comparison will help you navigate the trade-offs. By the end, you might find yourself questioning not just which model is better—but what 'better' really means in the evolving landscape of AI-driven development. GPT 5 vs Claude Sonnet Key Features of ChatGPT 5 and Claude Sonnet GPT-5, OpenAI's latest offering, is engineered for advanced reasoning and adaptability across diverse challenges. It employs a routing mechanism to optimize task performance, making it a versatile tool for developers. Claude Sonnet, a competing model, is recognized for its speed and visually refined outputs, catering to developers who prioritize efficiency and design aesthetics. While both models aim to assist in coding and problem-solving, their approaches and results differ significantly. Performance in Building a Multiplayer Tic-Tac-Toe Game To evaluate their coding capabilities, both models were tasked with creating a multiplayer tic-tac-toe game. The results highlighted distinct strengths and weaknesses: GPT-5: Delivered a functional and robust solution, though its interface was visually basic and required additional design work. Delivered a functional and robust solution, though its interface was visually basic and required additional design work. Claude Sonnet: Produced a polished and visually appealing interface but occasionally included unnecessary schema fields, complicating the codebase. Both models encountered minor coding errors and redundant schema elements, emphasizing the importance of manual refinement regardless of the model used. These findings suggest that while both tools are capable, neither can fully replace human oversight in development tasks. OpenAI ChatGPT 5 vs Claude Sonnet : Coding Showdown Watch this video on YouTube. Master AI coding with the help of our in-depth articles and helpful guides. Impact of Rule Adherence on Model Performance The ability to follow structured requirements, such as cursor guidelines and schema conventions, was tested to assess how rules influence performance. The models exhibited contrasting behaviors: GPT-5: Demonstrated slower but more deliberate processing, often providing thoughtful and detailed solutions that adhered closely to the rules. Demonstrated slower but more deliberate processing, often providing thoughtful and detailed solutions that adhered closely to the rules. Claude Sonnet: Executed tasks more quickly but struggled with schema conventions and index handling, leading to occasional inconsistencies in its outputs. These differences highlight how each model's inherent processing style affects its ability to meet structured requirements. While rules can improve performance, the quality of outputs depends heavily on the model's internal logic and reasoning capabilities. Challenges in Authentication Implementation Authentication tasks, such as implementing username/password and anonymous login systems, posed significant challenges for both models. The results revealed key insights: Both models faced difficulties with environment variable management and TypeScript compatibility, requiring additional manual intervention to resolve issues. GPT-5: Used web searches to compensate for its slower processing speed, often resulting in more complete and reliable solutions. Used web searches to compensate for its slower processing speed, often resulting in more complete and reliable solutions. Claude Sonnet: Executed tasks quickly but occasionally produced incomplete or less dependable functions, requiring further debugging. These findings underscore the complexity of authentication tasks and the need for developers to carefully review and refine outputs, regardless of the model used. Token Efficiency and Cost Implications Token efficiency is a critical factor in determining the cost-effectiveness of AI models. The comparison revealed distinct patterns in token usage: GPT-5 : Consumed more tokens due to its extensive reasoning process, often resulting in higher-quality outputs. Despite using more tokens, its lower per-token rates can make it cost-competitive, especially for complex projects where accuracy is essential. Claude Sonnet: Used fewer tokens but relied heavily on its context window, particularly when rules were not explicitly defined. However, its higher per-token rates can offset the savings from reduced usage. These differences highlight the trade-offs between reasoning depth and token efficiency. GPT-5 offers strong value for detailed, accuracy-driven work, while Claude Sonnet may be better suited for tasks that emphasize speed and brevity. Pricing and Cost Analysis Cost is a significant consideration when selecting an AI model. The pricing structures for GPT-5 and Claude Sonnet are as follows: GPT-5: $1.25 per million input tokens and $10 per million output tokens. $1.25 per million input tokens and $10 per million output tokens. Claude Sonnet: $3 per million input tokens and $15 per million output tokens. While GPT-5 is more affordable on a per-token basis, Claude Sonnet's lower overall token usage may still make it competitive for shorter, more focused tasks. The choice between the two models should align with your project's specific budget and requirements. Strengths and Weaknesses of Each Model The comparison highlights distinct strengths and weaknesses for GPT-5 and Claude Sonnet, making them suitable for different use cases: GPT-5: Strengths: Advanced reasoning, deliberate problem-solving, and high-quality outputs. Weaknesses: Slower processing, higher token usage, and occasional errors in execution. Claude Sonnet: Strengths: Faster execution, visually appealing outputs, and efficient token usage. Weaknesses: Inconsistent schema handling, less reliable methods, and occasional incomplete solutions. Choosing the Right Model for Your Needs Both GPT-5 and Claude Sonnet offer unique advantages, making them suitable for different types of projects. GPT-5 excels in tasks that require complex problem-solving and reasoning, though its slower processing and higher token usage may be drawbacks for time-sensitive projects. Claude Sonnet, on the other hand, provides faster execution and visually polished results but struggles with consistency and schema handling. Your choice should depend on your project's specific needs. If accuracy, reasoning, and detailed outputs are priorities, GPT-5 may be the better option. For tasks that demand speed and visual appeal, Claude Sonnet could be more suitable. As both models continue to evolve, ongoing testing and updates will provide deeper insights into their capabilities, allowing developers to make even more informed decisions in the future. Media Credit: Convex Filed Under: AI, Top News Latest Geeky Gadgets Deals Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.
