Latest AI News
Apple Has 'Complete Access' to Google's Gemini Model; Can Create Smaller Models via Distillation: Report
Apple has been granted full access to Google's Gemini model, which allows the iPhone maker to do more with the AI model used on Android smartphones, according to a report. The Cupertino company will be able to use the Gemini AI model for distillation in its own data centres, which means it can create smaller models that can be used for specific purposes. These models could be more efficient, would run on a user's device, and would not require access to the internet.
View
Yann LeCun Builds a World Model That Runs on a Single GPU
LeWorldModel can plan up to 48 times faster than some existing world models while maintaining competitive performance.
View
Infosys Announces FY26’s Largest Acquisition at $465 Mn; Total Deals Reach 5
The company has also acquired Stratus, extending the spree that includes Versent, MRE Consulting and The Missing Link.
View
Invention Engine Rewrites the Accelerator Playbook for Deep Tech
With smaller cohorts and operator-led guidance, the accelerator pushes founders to engage deeply with market realities rather than relying on theoretical frameworks.
View
How AlphaFold is Driving India’s Life Sciences Industry
This allows companies like GSK and Sanofi to speed up R&D by 30-40 per cent, enabling faster identification of drug targets
View
How Google Used High School Math to Deliver 8x Performance Boost on NVIDIA H100s
“All you had to do was pay attention to the polar coordinates lecture in [trigonometry], and you could have discovered a 6x reduction in KV cache memory.”
View
Mercor competitor Deccan AI raises $25M, sources experts from India
As demand grows for training and refining AI models,Deccan AI— a startup supplying post-training data and evaluation work — has raised $25 million in its first major funding round, with much of that work carried out by an India-based workforce of experts. The all-equity Series A round was led by A91 Partners, with participation from Susquehanna International Group and Prosus Ventures. While frontier AI labs including OpenAI and Anthropic build core models in-house, much of the post-training work — from data generation to evaluation and reinforcement learning — is increasingly being outsourced as companies push to make systems reliable in real-world use. Deccan is emerging as one of a new set of startups serving that demand. Founded in October 2024, Deccan provides services ranging from helping models improve coding and agent capabilities to training systems to interact with external tools such as application programming interfaces (APIs), which connect AI models to software systems. The startup works with frontier labs on tasks such as generating expert feedback, running evaluations and building reinforcement learning environments, while also serving enterprises through products including its evaluation suite, Helix, and an operations automation platform. The work is also evolving as models move beyond text into so-called “world models” that better understand physical environments, including robotics and vision systems. Deccan’s customers include Google DeepMind and Snowflake, according to the company. It has onboarded about 10 customers and runs a couple of dozen active projects at any given time, founder Rukesh Reddy (pictured above) said in an interview. The startup, headquartered in the San Francisco Bay Area with a large operations team in Hyderabad, employs about 125 people and relies on a network of more than 1 million contributors, including students, domain experts, and PhDs. Around 5,000 to 10,000 contributors are active in a typical month, Reddy told TechCrunch. About 10% of Deccan’s contributor base has advanced degrees such as master’s and PhDs, though the share is higher among active contributors depending on project requirements, Reddy said. The market for AI training services hasexpanded rapidlyalongside the rise of large language models, with companies such asMeta-owned Scale AIand its rivalSurge AI, as well as startupsTuringandMercorcompeting to provide data labeling, evaluation, and reinforcement learning services. “Quality remains an unsolved problem,” Reddy said, adding that tolerance for errors in post-training is “close to zero” as mistakes can directly affect model performance in production. That makes post-training more complex than earlier stages, requiring highly accurate, domain-specific data that is harder to scale. The work is also highly time-sensitive, he said, with AI labs sometimes requiring large volumes of high-quality data within days, making it difficult to balance speed with accuracy. The sector hasfaced criticism over working conditions and pay, with large pools of gig workers often used to generate training data. Reddy said earnings on Deccan’s platform range from about $10 to $700 per hour, with top contributors earning up to $7,000 a month. Even as its customers are largely U.S.-based AI labs, most of Deccan’s contributors are based in India. Competitors such as Turing and Mercoralso source contractorsfrom the country, but operate across abroader set of emerging markets. Deccan chose to concentrate much of its workforce in India to better manage quality, Reddy said. “Many of our competitors go to 100-plus countries to find the experts,” he said. “If you have operations in just one country, it becomes far easier to maintain quality.” That approach highlights India’s current position in the global AI value chain — as a supplier of talent and training data rather than a developer of frontier models, which remain concentrated among a handful of U.S. companies and a few players in China. However, Reddy said Deccan has begun sourcing talent from a few other markets, including the U.S., for niche expertise in geospatial data and semiconductor design. Reddy said Deccan was built as a “born GenAI” company, in contrast to traditional data labeling firms that began with computer vision tasks. This means it has focused on higher-skill work from the outset. Deccan grew 10x over the past year and is now at a double-digit million-dollar revenue run rate, Reddy said, declining to share specifics. About 80% of its revenue comes from its top five customers, reflecting the concentrated nature of the frontier AI market, he added.
View
The least surprising chapter of the Manus story is what’s happening right now
Okay, so the U.S. and China are locked in an all-out race to build the most powerful AI on the planet. Beijing is throwing billions at homegrown models, tightening its grip on the tech sector, and watching nervously as its best AI talentgravitates to U.S. companies. A Carnegie Endowment study published late last year found that 87 of the 100 top Chinese AI researchers at U.S. institutions in 2019 are still there. Yet Manus — one of China’s most buzzed-about AI startups — quietly relocated to Singapore and sold itself to Meta for $2 billion. Did anyone think there wouldnotbe a reckoning over this tie-up? As industry watchers know, Manus burst onto the scene in the spring of last year with a demo video showing an AI agent screening job candidates, planning vacations, and analyzing stock portfolios, and it cheekily claimed it outperformed OpenAI’s Deep Research. Within weeks, Benchmark — the consummate Silicon Valley venture firm — led a $75 million funding round at a $500 million valuation. That was surprising. (Senator John Cornyn had thoughts,tweetingat the time, “Who thinks it is a good idea for American investors to subsidize our biggest adversary in AI, only to have the CCP use that technology to challenge us economically and militarily? Not me.”) By December, Manus had millions of users and was pulling in over $100 million in annual recurring revenue. Then Meta came calling, and Mark Zuckerberg, who has staked the company’s future on AI, snapped it up for $2 billion. It’s worth noting that Manus didn’t just sell itself to an American buyer; it spent the better part of last year actively trying to operate outside China’s orbit. The company relocated its headquarters and core team from Beijing to Singapore, restructured its ownership, and after the Meta deal was announced, Metapledged to cut all tieswith Manus’s Chinese investors and shut down its operations in China entirely. By every measure, Manus was trying to make itself a Singapore company. But if that string of events raised eyebrows in Washington, you can only imagine that in Beijing, they were apoplectic. China has a phrase for all of this: “selling young crops” — homegrown AI companies that move abroad and sell themselves to foreign buyers before they’ve fully matured, taking their intellectual property and talent with them. Beijing hates it and has spent years establishing that no company operates outside its reach. Surely, we all remember that time Jack Ma gave a speech in 2020, mildly criticizing Chinese regulators, after which he disappeared from public life for months, Ant Group’s blockbuster IPO was killed overnight, and Alibaba was handed a $2.8 billion fine. China then spent the next two years methodically dismantling its own booming tech sector, wiping out hundreds of billions in market value. Chinese leaders are many things, but subtle is not one of them. Which is why it wasn’t entirely surprising when, on Tuesday, the Financial Times reported that Manus co-founders Xiao Hong and Ji Yichao were summoned to a meeting this month with China’s National Development and Reform Commission and told that theywouldn’t be leaving the countryfor a while. No formal charges have been filed — just an inquiry into whether the Meta deal violated Beijing’s foreign investment rules. Beijing is calling it a routine regulatory review. At some point, someone at Manus probably thought they’d gotten away with it, and maybe they still will. But given the stakes of the AI race, that was always a big gamble. Now Beijing wants answers; Manus’s founders are apparently not going anywhere until it gets them.
View
Melania Trump wants a robot to homeschool your child
At apress conferenceat the White House on Wednesday, First Lady Melania Trump showed up with a humanoid robot developed by robotics firm Figure AI. The duo waltzed down a red carpet together before the bot gave a brief speech, chirping: “I am grateful to be part of this historic movement to empower children with technology and education.” Not long after these remarks, the machine mosied out of the room and disappeared. The bizarre spectacle was part of the first lady’s newly launched initiative, theFostering the Future Togetherglobal summit, which invited international leaders from around the world to discuss how to empower children through educational technology, including AI. The event easily evoked dystopian visions of the future — ones in which the humble (human) school teacher has been replaced by a Terminator-shaped server stack that can walk and speak Latin. Indeed, during her remarks, the first lady asked attendees to imagine a future in which a humanoid robot would act as the ultimate educator for the world’s children. (The event took place at the same time the Trump administrationannounced a separate tech councilstaffed by jet-setting Silicon Valley executives.) “Imagine a humanoid educator named Plato,” thefirst lady said. “Access to the classical studies is now instantaneous — literature, science, art, philosophy, mathematics, and history — Humanity’s entire corpus of information is available in the comfort of your home. Plato will provide a personalized experience, adaptive to the needs of each student. Plato is always patient, and always available. Predictably, our children will develop deeper critical thinking and independent reasoning abilities.” “Honored to be invited to the White House by the First Lady Melania Trump,” the Figure AIX account postedon Wednesday. The first lady’s comments are obviously forward-looking and don’t reflect where robotics and edtech are today, or will be anytime soon. Still, the thinking that AI and technology can be used to automate learning (and, in many ways, replace human educators) has beengaining in popularityin the tech industry. Such ideas have been repeatedly promoted by the White House. Over the past year, educational experiments like the Alpha School, a network of private schools thatuse AI to teach childrenat a rapid speed, have gained traction and media attention. The Trump administration has embraced experiments like these, while simultaneously attacking the traditional public education system. Secretary of Education Linda E. McMahon, who is in the midst ofabolishing the very agencyshe is tasked with running, has also found the time to visit an Alpha School campus, where she recently praised the “opportunity” promised by the educational chain. “Alpha School is reimagining K–12 education by equipping students with practical AI skills and preparing them for a rapidly evolving technology-driven workforce,” the administrationrecently saidof McMahon’s visit. Melania Trump’s event Wednesday similarly highlighted the role that the administration feels the tech industry should play in the future of American education — withTrump recognizingthe “participation of leading American technology companies, whose engagement reflects the growing role of the private sector in supporting safe and effective educational innovation.” Honored to be invited to the White House by the First Lady Melania Trumppic.twitter.com/E8J74hOciq
View
Google unveils TurboQuant, a new AI memory compression algorithm — and yes, the internet is calling it ‘Pied Piper’
If Google’s AI researchers had a sense of humor, they would have calledTurboQuant, the new, ultra-efficient AI memory compression algorithm announced Tuesday, “Pied Piper” — or, atleastthat’swhattheinternetthinks. The joke is a reference to the fictional startup Pied Piper that was the focus of HBO’s “Silicon Valley” TV series that ran from 2014 to 2019. The show followed the startup’s founders as they navigated the tech ecosystem, facing challenges like competition from larger companies, fundraising, technology and product issues, and even (much to our delight)wowing the judges at a fictional version ofTechCrunch Disrupt. Pied Piper’s breakthrough technology on the TV show was a compression algorithm that greatly reduced file sizes with near-lossless compression. Google Research’s newTurboQuantis also about extreme compression without quality loss, but applied to a core bottleneck in AI systems. Hence, the comparisons. So Google TurboQuant is basically Pied Piper and just hit a Weismann Score of 5.2https://t.co/WievkwijjDpic.twitter.com/4rirvu2YyV Google Researchdescribed the technologyas a novel way to shrink AI’s working memory without impacting performance. The compression method, which uses a form of vector quantization to clear cache bottlenecks in AI processing, would essentially allow AI to remember more information while taking up less space and maintaining accuracy, according to the researchers. They plan to present their findings at theICLR 2026conference next month, along with the two methods that are making this compression possible: the quantization methodPolarQuantand a training and optimization method calledQJL. TurboQuant is the new Pied Piper 🤣pic.twitter.com/iMAYJs02zt So basically TurboQuant is Pied Piperhttps://t.co/Zx9Oq84tSLpic.twitter.com/JPZjz8M3Wp Understanding the math involved here is something researchers and computer scientists may be able to do, but the results are exciting the wider tech industry as a whole. If successfully implemented in the real world, TurboQuant could make AI cheaper to run by reducing its runtime “working memory” — known as the KV cache — by “at least 6x.” Some, like Cloudflare CEO Matthew Prince, areeven calling thisGoogle’sDeepSeek moment— a reference to theefficiency gainsdriven by the Chinese AI model, which was trained at a fraction of the cost of its rivals on worse chips, while remaining competitive on its results. This is Google’s DeepSeek. So much more room to optimize AI inference for speed, memory usage, power consumption, and multi-tenant utilization. Lots of teams at@Cloudflarefocused on these areas.#staytunedhttps://t.co/hHoY4sLT2I Well, we all know who stole the Pied Piper codebase nowhttps://t.co/Inv0nlMYnP Still, it’s worth noting that TurboQuant hasn’t yet been deployed broadly; it’s still a lab breakthrough at this time. That makes comparisons with something like DeepSeek, or even the fictional Pied Piper, more difficult. On TV, Pied Piper’s technology was going to radically change the rules of computing. TurboQuant, meanwhile, could lead to efficiency gains and systems that require less memory during inference. But it wouldn’t necessarily solve the wider RAM shortages driven by AI, given that it only targets inference memory, not training — the latter of which continues to require massive amounts of RAM. Pied Piper would have been a better namehttps://t.co/qNZmtANFhs
View
The AI skills gap is here, says AI company, and power users are pulling ahead
Anthropic’slatest researchsuggests that while AI is rapidly changing the way work gets done, it hasn’t meaningfully eliminated jobs. At least, not yet. But beneath what Anthropic’s head of economics, Peter McCrory, says is a “still healthy” labor market, early signs are pointing to uneven impacts, especially for younger workers just entering the workforce. In an interview on the sidelines of the Axios AI Summit in Washington, D.C., McCrory said the company’s newest economic impact report finds little evidence of widespread job displacement so far. “There’s no material difference in unemployment rates” between workers who use Claude for the “most central task of their job in automated ways” — like technical writers, data entry clerks, and software engineers — and workers in jobs less exposed to AI that require “physical interaction and dexterity with the real world.” But with AI adoption spreading across industries, that could shift — fast. If Anthropic CEO Dario Amodei is to be believed, AI couldwipe out halfof all entry-level white-collar jobs and push unemployment as high as 20% within the next five years. @PeterMcCroryis@AnthropicAI’s head of economics. He says there are no real signs of AI-related job loss yet.That doesn’t mean it’s not coming.Reporting from@axiosAI summit in DC. https://t.co/sScZDC43o7 “Displacement effects could materialize very quickly, so you want to establish a monitoring framework to understand that before it materializes so that we can catch it as it’s happening and ideally identify the appropriate policy response,” McCrory told TechCrunch. Staying ahead of those trends is whytrackingAI growth, adoption, and diffusion is so important, he said. In theory, McCrory said, AI models like Claude can do almost anything a computer can do. In practice, most users are only scratching the surface of those capabilities. He said Anthropic looked at which roles involve tasks that AI is particularly good at, that are already being automated, and that are tied to real workplace use cases — the areas most likely to signal where displacement could emerge. Anthropic’s fifth economic impact report, released Tuesday, also found that even where there hasn’t been much displacement yet, there’s a growing skills gap between earlier Claude adopters and newcomers. Earlier adopters are more likely to get significantly more value from the model, using it for work-related tasks rather than casual or one-off purposes and in more sophisticated ways, like as a “thought partner” for iteration and feedback. McCrory said the findings suggest AI is becoming a technology that rewards those who already know how to use it — and that workers who can effectively incorporate it into their work will increasingly have an edge. That advantage isn’t evenly distributed geographically, either. The report also found that “Claude is used more intensely in high-income countries, within the U.S. in places with more knowledge workers, and for a relatively small set of specialized tasks and occupations.” In other words, despite promises of AI as an equalizer, adoption may already be tilting toward the wealthy and could amplify those advantages as power users pull further ahead.
View
Harvey confirms $11B valuation: Sequoia triples down
One of the blockbuster hits of the AI age is, without a doubt,legal tech startup Harvey. On Wednesday, the companyconfirmedthat it had closed a new raise at an $11 billion valuation, after reportscirculated last monththat it was working on another monster round. The company confirmed it inhaled $200 million from this round, co-led by returning investors Singapore’s GIC and Sequoia. Existing investors Andreessen Horowitz, Coatue, Conviction Partners, Elad Gil, Evantic, and Kleiner Perkins also participated. With this new funding, the company has raised more than $1 billion in total, and its valuation jumped over 3.5x in a year. Harvey was valued at $8 billion from a round announced in December, led by Andreessen Horowitz. Before that, it was valued at $5 billion from a round led by Kleiner Perkins and Coatue, announced in June, and was at $3 billion from a Sequoia-led raise announced in February 2025. Sequoia has now co-led three of its rounds since its Series A, a move even Sequoia partner Pat Grady acknowledged was an unusually large show of faith for the VC firm, Grady said in the press release. A few months ago, founder and CEO Winston Weinbergdescribed to TechCrunch’s EIC what a wild ride it’s been.
View
