Nvidia-backed startup CEO urges Japan to push defense AI development
Agencies
The top executive of a leading Japanese artificial intelligence startup said the Japanese government should focus more on the development of its own AI technology for defense applications in an increasingly deglobalized world.
David Ha, the chief executive of Sakana AI KK, backed by investors such as U.S. chip giant Nvidia Corp, said his company sees growth potential in working with the Japanese government in the defense sector since the country's security environment is becoming severe with 'many adversaries around its neighbors.'
'Not many startups in Japan want to work in (the) defense side and (with) governments,' Ha said at a recent press conference at the Foreign Correspondents' Club of Japan.It presents 'an opportunity for our company to engage with the governments on developing defense-related solutions using AI,' said the CEO at the Tokyo-based startup known as a unicornan unlisted company valued at more than $1 billion.
Technology plays a key role in bolstering information technology infrastructure, and AI is a core part of such technology to make information systems and cyber security more secured and efficient, the former researcher at Google LLC said.
Since the United States has become more U.S.-centric, there is a potential for AI to become 'a bargaining chip for other countries,' if its services and models are so powerful, Ha said.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Qatar Tribune
20 hours ago
- Qatar Tribune
Europe seeks ‘digital sovereignty' as US tech firms embrace Trump
Agencies At a market stall in Berlin run by charity Topio, volunteers help people who want to purge their phones of the influence of US tech firms. Since Donald Trump's inauguration, the queue for their services has grown. Interest in European-based digital services has jumped in recent months, data from digital market intelligence company Similarweb shows. More people are looking for e-mail, messaging and even search providers outside the United States. The first months of Trump's second presidency have shaken some Europeans' confidence in their long-time ally, after he signaled his country would step back from its role in Europe's security and then launched a trade war. 'It's about the concentration of power in US firms,' said Topio's founder Michael Wirths, as his colleague installed on a customer's phone a version of the Android operating system without hooks into the Google ecosystem. Wirths said the type of people coming to the stall had changed: 'Before, it was people who knew a lot about data privacy. Now it's people who are politically aware and feel exposed.' Tesla chief Elon Musk, who also owns social media company X, was a leading adviser to the US president before the two fell out, while the bosses of Amazon, Meta and Google-owner Alphabet took prominent spots at Trump's inauguration in January. Days before Trump took office, outgoing president Joe Biden had warned of an oligarchic 'tech industrial complex' threatening democracy. Berlin-based search engine Ecosia says it has benefited from some customers' desire to avoid US counterparts like Microsoft's Bing or Google, which dominates web searches and is also the world's biggest email provider. 'The worse it gets, the better it is for us,' founder Christian Kroll said of Ecosia, whose sales pitch is that it spends its profits on environmental projects. Similarweb data shows the number of queries directed to Ecosia from the European Union has risen 27 percent year-on-year and the company says it has 1 percent of the German search engine market. But its 122 million visits from the 27 EU countries in February were dwarfed by 10.3 billion visits to Google, whose parent Alphabet made revenues of about $100 billion from Europe, the Middle East and Africa in 2024 - nearly a third of its $350 billion global turnover. Non-profit Ecosia earned 3.2 million euros ($3.65 million) in April, of which 770,000 euros was spent on planting 1.1 million trees. Google declined to comment for this story. Reuters could not determine whether major US tech companies have lost any market share to local rivals in Europe. The search for alternative providers accompanies a debate in Europe about 'digital sovereignty' - the idea that reliance on companies from an increasingly isolationist United States is a threat to Europe's economy and security. 'Ordinary people, the kind of people who would never have thought it was important they were using an American service are saying, 'hang on!',' said UK-based internet regulation expert Maria Farrell. 'My hairdresser was asking me what she should switch to.' Use in Europe of Swiss-based ProtonMail rose 11.7 percent year-on-year to March compared to a year ago, according to Similarweb, while use of Alphabet's Gmail, which has some 70 percent of the global email market, slipped 1.9 percent. ProtonMail, which offers both free and paid-for services, said it had seen an increase in users from Europe since Trump's re-election, though it declined to give a number. 'My household is definitely disengaging,' said British software engineer Ken Tindell, citing weak US data privacy protections as one factor. Trump's vice president JD Vance shocked European leaders in February by accusing them - at a conference usually known for displays of transatlantic unity - of censoring free speech and failing to control immigration. In May, Secretary of State Marco Rubio threatened visa bans for people who 'censor' speech by Americans, including on social media, and suggested the policy could target foreign officials regulating US tech companies. US social media companies like Facebook and Instagram parent Meta have said the European Union's Digital Services Act amounts to censorship of their platforms. EU officials say the Act will make the online environment safer by compelling tech giants to tackle illegal content, including hate speech and child sexual abuse material. Greg Nojeim, director of the Security and Surveillance Project at the Center for Democracy & Technology, said Europeans' concerns about the US government accessing their data, whether stored on devices or in the cloud, were justified. Not only does US law permit the government to search devices of anyone entering the country, it can compel disclosure of data that Europeans outside the US store or transmit through US communications service providers, Nojeim said. Germany's new government is itself making efforts to reduce exposure to US tech, committing in its coalition agreement to make more use of open-source data formats and locally-based cloud infrastructure. Regional governments have gone further - in conservative-run Schleswig-Holstein, on the Danish border, all IT used by the public administration must run on open-source software. Berlin has also paid for Ukraine to access a satellite-internet network operated by France's Eutelsat instead of Musk's Starlink. But with modern life driven by technology, 'completely divorcing US tech in a very fundamental way is, I would say, possibly not possible,' said Bill Budington of US digital rights nonprofit the Electronic Frontier Foundation. Everything from push notifications to the content delivery networks powering many websites and how internet traffic is routed relies largely on US companies and infrastructure, Budington noted. Both Ecosia and French-based search engine Qwant depend in part on search results provided by Google and Microsoft's Bing, while Ecosia runs on cloud platforms, some hosted by the very same tech giants it promises an escape from. Nevertheless, a group on messaging board Reddit called BuyFromEU has 211,000 members. 'Just cancelled my Dropbox and will switch to Proton Drive,' read one post. Mastodon, a decentralized social media service developed by German programmer Eugen Rochko, enjoyed a rush of new users two years ago when Musk bought Twitter, later renamed X. But it remains a niche service.
Qatar Tribune
4 days ago
- Qatar Tribune
Beyond the internet: AI learning from 15th-century texts
Agencies Everything ever said on the internet was just the start of teaching artificial intelligence about humanity. Tech companies are now tapping into an older repository of knowledge: the library stacks. Nearly one million books published as early as the 15th century — and in 254 languages — are part of a Harvard University collection being released to AI researchers Thursday. Also coming soon are troves of old newspapers and government documents held by Boston's public library. Cracking open the vaults to centuries-old tomes could be a data bonanza for tech companies battling lawsuits from living novelists, visual artistsand others whose creative works have been scooped up without their consent to train AI chatbots. 'It is a prudent decision to start with public domain data because that's less controversial right now than content that's still under copyright,' said Burton Davis, a deputy general counsel at Microsoft. Davis said libraries also hold 'significant amounts of interesting cultural, historical and language data' that's missing from the past few decades of online commentary that AI chatbots have mostly learned from. Supported by 'unrestricted gifts' from Microsoft and ChatGPT maker OpenAI, the Harvard-based Institutional Data Initiative is working with libraries around the world on how to make their historic collections AI-ready in a way that also benefits libraries and the communities they serve. 'We're trying to move some of the power from this current AI moment back to these institutions,' said Aristana Scourtas, who manages research at Harvard Law School's Library Innovation Lab. 'Librarians have always been the stewards of data and the stewards of information.' Harvard's newly released dataset, Institutional Books 1.0, contains more than 394 million scanned pages of paper. One of the earlier works is from the 1400s — a Korean painter's handwritten thoughts about cultivating flowers and trees. The largest concentration of works is from the 19th century, on subjects such as literature, philosophy, law and agriculture, all of it meticulously preserved and organized by generations of librarians. It promises to be a boon for AI developers trying to improve the accuracy and reliability of their systems. 'A lot of the data that's been used in AI training has not come from original sources,' said the data initiative's executive director, Greg Leppert, who is also chief technologist at Harvard's Berkman Klein Center for Internet & Society. This book collection goes 'all the way back to the physical copy that was scanned by the institutions that actually collected those items,' he said. Before ChatGPT sparked a commercial AI frenzy, most AI researchers didn't think much about the provenance of the passages of text they pulled from Wikipedia, from social media forums like Reddit and sometimes from deep repositories of pirated books. They just needed lots of what computer scientists call tokens — units of data, each of which can represent a piece of a word. Harvard's new AI training collection has an estimated 242 billion tokens, an amount that's hard for humans to fathom but it's still just a drop of what's being fed into the most advanced AI systems. Facebook parent company Meta, for instance, has said the latest version of its AI large language model was trained on more than 30 trillion tokens pulled from text, images and videos. Meta is also battling a lawsuit from comedian Sarah Silverman and other published authors who accuse the company of stealing their books from 'shadow libraries' of pirated works. Now, with some reservations, the real libraries are standing up. OpenAI, which is also fighting a string of copyright lawsuits, donated $50 million this year to a group of research institutions including Oxford University's 400-year-old Bodleian Library, which is digitizing rare texts and using AI to help transcribe them. When the company first reached out to the Boston Public Library, one of the biggest in the U.S., the library made clear that any information it digitized would be for everyone, said Jessica Chapel, its chief of digital and online services. 'OpenAI had this interest in massive amounts of training data. We have an interest in massive amounts of digital objects. So this is kind of just a case that things are aligning,' Chapel said. Digitization is expensive. It's been painstaking work, for instance, for Boston's library to scan and curate dozens of New England's French-language newspapers that were widely read in the late 19th and early 20th century by Canadian immigrant communities from Quebec. Now that such text is of use as training data, it helps bankroll projects that librarians want to do anyway. 'We've been very clear that, 'Hey, we're a public library,'' Chapel said. 'Our collections are held for public use, and anything we digitized as part of this project will be made public.' Harvard's collection was already digitized starting in 2006 for another tech giant, Google, in its controversial project to create a searchable online library of more than 20 million books. Google spent years beating back legal challenges from authors to its online book library, which included many newer and copyrighted works. It was finally settled in 2016 when the U.S. Supreme Court let stand lower court rulings that rejected copyright infringement claims. Now, for the first time, Google has worked with Harvard to retrieve public domain volumes from Google Books and clear the way for their release to AI developers. Copyright protections in the U.S. typically last for 95 years, and longer for sound recordings. How useful all of this will be for the next generation of AI tools remains to be seen as the data gets shared Thursday on the Hugging Face platform, which hosts datasets and open-source AI models that anyone can download. The book collection is more linguistically diverse than typical AI data sources. Fewer than half the volumes are in English, though European languages still dominate, particularly German, French, Italian, Spanish and Latin. A book collection steeped in 19th century thought could also be 'immensely critical' for the tech industry's efforts to build AI agents that can plan and reason as well as humans, Leppert said. 'At a university, you have a lot of pedagogy around what it means to reason,' Leppert said. 'You have a lot of scientific information about how to run processes and how to run analyses.' At the same time, there's also plenty of outdated data, from debunked scientific and medical theories to racist narratives. 'When you're dealing with such a large data set, there are some tricky issues around harmful content and language,' said Kristi Mukk, a coordinator at Harvard's Library Innovation Lab who said the initiative is trying to provide guidance about mitigating the risks of using the data, to 'help them make their own informed decisions and use AI responsibly.'
Qatar Tribune
4 days ago
- Qatar Tribune
Slump in auto exports amid tariffs hits Japan's shipments to United States
Agencies Japan's exports dropped in May for the first time in eight months as top automakers like Toyota were hit by sweeping U.S. tariffs – while Tokyo also did not manage to strike a trade deal with Washington this week – which would likely put even more pressure on a fragile economy. Japan's Prime Minister Shigeru Ishiba said after the G-7 summit in Canada on Tuesday that his country had not reached a comprehensive tariff agreement with Washington, as some disagreements persisted between the two nations despite several rounds of talks. Japan and the U.S. 'explored the possibility of a deal until the last minute,' he added. Tokyo is scrambling to find ways to get Washington to exempt Japan's automakers from 25% automobile industry-specific tariffs, which are hurting the country's manufacturing sector. Japan also faces a 24% 'reciprocal' tariff rate starting on July 9 unless it can negotiate a deal with Washington. The data on Wednesday showed that Japanese auto exports to the U.S. fell almost a quarter in May as worries over tariffs grow. Roughly 8% of jobs are tied to the auto industry in Japan, which is home to the world's top-selling carmaker, Toyota, as well as Honda, Nissan and other giants. Japan's automobile sector accounted for about 28% of the total 21 trillion yen ($145 billion) worth of goods the Asian country exported to the U.S. last year. Its total exports in May dropped 1.7% year-over-year by value to 8.1 trillion yen, government data showed, smaller than a median market forecast for a 3.8% decrease, and following a 2% rise in April. Exports to the U.S. slumped 11.1% last month from a year earlier, the largest monthly percentage decline since February 2021, dragged down by a 24.7% plunge in automobiles and a 19% fall in auto components, while a stronger yen also helped reduce the value of shipments. Exports to China were down 8.8%. In terms of volume, however, U.S.-bound automobile exports dipped just 3.9%, indicating that the biggest Japanese exporters were absorbing the tariff costs. 'The value of automobile exports to the U.S. fell, but their volume did not drop that much,' Daiwa Institute of Research economist Koki Akimoto said. 'This indicates Japanese automakers are effectively shouldering the tariff costs and not charging customers.' So far, major Japanese automakers have refrained from price increases in the U.S. to mitigate the tariff costs, except for Subaru and Mitsubishi Motors. 'They are buying time right now to see the course of Japan-U.S. trade negotiations,' Akimoto said. The absence of price hikes could affect their profits, but their fiscal base is generally solid, he added. While Japanese stocks and the yen showed little reaction to the data, shares of car companies have come under pressure this year due to concerns about the tariff impact. Automakers and other transport companies are the second-worst performers this year among the Tokyo market's 33 sector sub-indices, down almost 12%. Only makers of precision equipment have fared worse. Toyota, the world's top-selling automaker, has estimated that tariffs likely sliced 180 billion yen from its profit in April and May alone. Honda has said it expects a 650 billion yen hit to its earnings this year from tariffs in the U.S. and elsewhere. The Japan May trade data provide one of the earliest indications of how U.S. President Donald Trump's tariffs are impacting countries and the global economy. China's data showed this week that the country's factory output grew 5.8% in May year-over-year, the slowest pace in six months. And its outbound shipments to the U.S. plunged 34.5%, the sharpest drop since February 2020. The impending tariffs had driven companies in Japan and other major Asian exporters to ramp up shipments earlier this year, inflating levels of U.S.-bound exports during that period. The Japan data showed imports dropped 7.7% in May from a year earlier, compared with market forecasts for a 6.7% decrease. As a result, Japan ran a trade deficit of 637.6 billion yen last month, compared with the forecast of a deficit of 892.9 billion yen. After the G-7 summit in Canada, Ishiba told reporters that U.S. tariffs were 'hitting many Japanese companies' profits.' The situation 'could have a grave impact on both Japan and the U.S. as well as the world economy, directly and indirectly,' he warned. The hit from U.S. tariffs could add more pressure on Japan's lackluster economy. Subdued private consumption already caused the world's fourth-largest economy to shrink in January-March, the first contraction in a year. However, the smaller-than-expected drop in May shipments suggests that Japan's export driver has not stumbled, slightly raising the chance of the economy avoiding a contraction in the April-June quarter, Yuhi Kawano, economist at Mizuho Securities, wrote in a report. The tariff woes complicate the Bank of Japan's (BOJ) task of raising still-low interest rates and reducing a balance sheet that has ballooned to roughly the size of Japan's economy. The BOJ kept interest rates steady on Tuesday and decided to decelerate the pace of its balance sheet drawdown next year, signaling its preference to move cautiously in removing remnants of its massive, decadelong stimulus. According to an estimate by the Japan Research Institute, if all the threatened tariff measures against Japan were to take effect, U.S.-bound exports would fall by 20% to 30%. Some economists say those duties could shave around one percentage point of the nation's gross domestic product (GDP).
