China is building the future of AI, not Silicon Valley, says Alibaba Cloud founder
In an interview with Bloomberg, Wang said Chinese foundational AI models like Qwen and DeepSeek are much better than OpenAI's ChatGPT, adding that China is a testbed for new technology.
'Foundational models like Qwen and DeepSeek are much better than ChatGPT. So we really need to fund creative people to build applications for them. In terms of applications, we are heavily biased toward OpenAI, because everyone sees ChatGPT as the only application that can provide security,' he said. 'The Chinese market has a very important role in establishing new technology and making sure it is mature enough, positioning the country as a testbed of every new technology to get products to market,' he added.
OpenAI vs Alibaba When asked about the stiff competition in the AI space, Wang said it's no less than a marathon for new players to enter the AI race, adding that healthy competition enables fast replication of the technology.'When people get together and it is not just for competition, whether you win or not, you can have a very fast iteration of the technology because of the competition.'Commenting on Silicon Valley's progress on building AI capabilities, he said just a single organisation, or individual, cannot go far in this journey. Additionally, he said that China is a country that benefits from a stable mindset. To make his point, Wang cited the example of Hangzhou, a city in China, claiming that one out of every four or five people there is a 'CEO.'
Poaching war: Meta, OpenAI Wang also addressed the big pay packets being offered in Silicon Valley to hire AI talent.According to him, the driving force for any organisation should be innovation, not patents.'What's happening in Silicon Valley is not the winning formula. We need the right talent, not expensive talent,' he said.'When you are in the early stage of innovation, I don't think a patent is a problem because the only thing you need to do is to get the right person, not really an expensive person,' he added.
ET reported recently that over a dozen staff at Mira Murati's AI startup, Thinking Machines Lab (TML), have been approached or offered jobs by Meta.
This talent poaching follows a previous instance reported when Meta hired four AI researchers from OpenAI. The tech giant has bagged top talent from companies, including OpenAI, Anthropic, and GitHub, after top-level exits and a poor reception for its latest open-source Llama 4 model. Following the trend, the Sam Altman-led OpenAI also poached four top engineers from rival firms led by Elon Musk and Mark Zuckerberg last month. Elevate your knowledge and leadership skills at a cost cheaper than your daily tea. US tariff hike to hit Indian exports, may push RBI towards rate cuts
Is Bajaj Finance facing its HDFC Bank moment?
Tata Motors' INR38k crore Iveco buy: Factors that can make investors nervous
Trump tariffs: End of road or a new journey ending Russia reliance?
Stock Radar: PI Industries stock showing signs of momentum; takes support above 50-DEMA – time to buy?
Long-term investing: Volatility, even threats, have limited shelf life; 5 large-caps from different sectors with upside potential of up to 38%
These large- and mid-cap stocks can give more than 21% return in 1 year, according to analysts
Weekly Top Picks: These stocks scored 10 on 10 on Stock Reports Plus
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Hans India
15 minutes ago
- Hans India
GPT-5 Launch Set for Tonight: Sam Altman Teases Major Upgrades in OpenAI Livestream
OpenAI is preparing to make waves in the AI world yet again with the highly anticipated launch of ChatGPT-5, scheduled for tonight. The livestream is set to begin at 10:00 AM Pacific Time, which is 10:30 PM IST, and will be broadcast on OpenAI's official YouTube channel. While the event link is yet to go live, it's expected to be made available shortly before the stream begins. OpenAI may also share the event on its official X (formerly Twitter) page, based on its past event practices. Anticipation around this launch has been growing for months, particularly as OpenAI CEO Sam Altman has hinted at significant upcoming developments. He previously teased "a ton of stuff" in the pipeline — encompassing not only GPT-5 but also a host of new products and AI tools. With such statements building momentum, all eyes are now on what OpenAI will unveil. The arrival of GPT-5 has been delayed several times, but signs now point to a launch that was worth the wait. Leaks and hints suggest that GPT-5 will be a major evolution from its predecessor, GPT-4. Users can expect noticeable improvements in performance, including faster response times, better reasoning abilities, and advanced multimodal support — meaning the model might now handle not just text, but also images, voice, and possibly even video. Several versions of GPT-5 have already appeared on GitHub Models, hinting at a more modular and adaptive AI ecosystem. These include: GPT-5 for complex reasoning and multi-step problem-solving, for complex reasoning and multi-step problem-solving, GPT-5-mini and GPT-5-nano for lightweight, budget-friendly, and low-latency tasks, and for lightweight, budget-friendly, and low-latency tasks, GPT-5-chat, a variant designed specifically for deep, contextual, and human-like conversations, especially for enterprise users. These variants are accessible via the GitHub Models API and playground, allowing users and developers to explore the new capabilities ahead of general rollout. Perhaps the most talked-about rumored feature is autonomous task handling. Instead of walking the model through every step of a task, users may soon be able to issue a single prompt and let GPT-5 take over from there — a significant leap toward true AI assistance. If implemented effectively, this would mark a dramatic shift in how users interact with AI for productivity, project management, and complex workflows. Another key improvement expected is enhanced memory. This would allow GPT-5 to recall past interactions, maintaining context over longer conversations and offering more personalized responses — a feature that could change the game for both casual users and businesses. While full access details and pricing tiers remain under wraps, speculation continues around whether OpenAI will open-source any part of GPT-5 or restructure its free and paid subscription models. With all these enhancements potentially bundled into a single, more unified experience, GPT-5 is poised to redefine how AI integrates into our digital lives. Stay tuned as OpenAI goes live with its biggest update yet.
The Hindu
15 minutes ago
- The Hindu
Trump plans 100% tariff on chips made outside U.S.; OpenAI's GPT-5 nears release; Apple to invest $100 billion in U.S. manufacturing
Trump plans 100% tariff on chips made outside U.S. U.S. President Donald Trump said that he will be imposing 100% tariff on computer chips unless they're manufactured within the U.S. The threat has raised the fears around higher prices of electronics, automobiles, household appliances and other essential electronic items. Trump's warning comes three months after he had said electronics would mostly be exempt from the heavy tariffs imposed by his administration. During the COVID-19 pandemic, a shortage of chips pushed up the price of autos causing increased inflation. Investors have said that the tariff exemptions were a plus for Apple and other major tech companies which have invested massive amounts of money in chip manufacturing outside the U.S. Apple has committed to investing $600 billion in the U.S. with another $100 billion promised in February. Big Tech in total has said they will be investing $1.5 trillion in the U.S. since Trump has come to power. There is a chance that Trump could exempt Apple from the tariffs owing to this. Demand for chips has grown with sales increasing by 19.6% in June, according to data. Trump has been fostering domestic manufacturing often saying that the high costs of chips outside the country could be avoided by opening factories domestically. OpenAI's GPT-5 nears release OpenAI is preparing for the release of their latest flagship AI model, GPT-5, so users can finally see the increments in performance from GPT-4. A couple of people who have tested the model said that the coding capabilities of the model shone as did its ability to solve science and math problems. However, the advances from GPT-4 to GPT-5 were not as huge as the advances from GPT-3 to GPT-4. The recent advancement in AI models was based on scaling up by increasing training data and compute. But OpenAI has struggled with scaling up like issues with running into a data wall. Ilya Sutskever, the former chief scientist at OpenAI had said in a speech earlier that even though power could be expanded, data was becoming limited. Besides this, another problem was that the 'training runs' for large language models are more likely to have failures from hardware malfunctioning so researchers can't determine the final performance of the models until the end which can last for months. OpenAI released ChatGPT around three years ago introducing AI to the mainstream globally. The release led to rivals like Google and Anthropic releasing competitive AI models. OpenAI is also working on solving test-time compute which involves reasoning AI models. But CEO Sam Altman said that GPT-5 will be a mix of both reasoning and their flagship large language models. Apple to invest $100 billion in U.S. manufacturing Apple CEO Tim Cook met Donald Trump yesterday to announce an additional investment of $100 billion across the next four years. Under the announcement, Apple will bring more of their supply chain and advanced manufacturing processes to the U.S. as a part of the American Manufacturing Programme initiative but it still isn't a full commitment to make the iPhone domestically completely. Cook said that this plan will expand work with 10 companies across the U.S. that make parts and semiconductor chips. These partners include Corning, Coherent, Applied Materials, Texas Instruments and Broadcom along with a few others. Apple had earlier this year said that it will invest $500 billion domestically. Trump has been upset with Apple shifting their manufacturing processes to India from China after he placed tariffs. Just recently now, Trump is targetting India due to their purchase of Russian oil and imposed additional tariffs on India. Cook has previously said that a bunch of iPhone components can be made in India like the glass display and module to identify faces.
The Hindu
15 minutes ago
- The Hindu
‘Unequal growth post the globalisation boom of the 1990s and 2000s led to the current inward looking trade policies'
The increasingly inward looking attitude of countries around the world didn't happen overnight, and is a result of the unequal growth that emerged from the 'golden' period of globalisation in the 1990s and 2000s, according to Scott Wang, Vice President, Asia Pacific, for the World Trade Centres' Association. Speaking to The Hindu, Mr. Wang acknowledged that the world is becoming more divided and that the international trade landscape was becoming more fragmented. However, he also noted that this was a transition period towards a more inclusive and well-rounded growth model. 'There is a rise of unilateralism, protectionism, and nationalism,' Mr. Wang explained. 'These are all sort of inward looking. Tariffs are the same thing. It's to protect domestic manufacturers.' He added that inward-looking doesn't necessarily mean focussing exclusively on domestic markets, but that international policy is also becoming increasingly region-based or block-based, resulting in trends like near-shoring and friend-shoring. However, he also explained that the current situation was not something that emerged suddenly, or as a result of any particular person. Rather, it is a result of the kind of globalisation that took place in the 1990s and 2000s. 'If you look at the situation right now, it didn't happen overnight,' Mr. Wang explained. 'It has kind of been evolving over the past decade at least.' 'When you're talking about the 'golden era' [the 1990s and 2000s] of globalisation, it brought prosperity and growth on a global scale, but there were also issues,' he added. 'The issue was that the distribution of wealth was very uneven among countries and among classes inside a country.' This, he said, began to create problems as people began to realise that the benefits of globalisation were accruing to a few. As a result, people began to use their votes and voices against this trend, which began to have a significant impact on trade policy and immigration laws. 'So, you see the rise of nationalism, sometimes you see anti-immigration, things like that,' he explained. 'These are all kind of because of some of the issues with that 'golden' version of globalisation. So, the World Trade Organisation's position is that we're moving towards a more inclusive, more sustainable, more common prosperity type of globalisation, but we're not there yet.' 'We're probably in a transitional period where you see a lot of uncertainty, a lot of turmoil,' he added. Mr. Wang also spoke about the annual Geopolitical Annual Trade Risk Index (GATRI) that has been developed by the WTCA and the Hague Centre for Strategic Studies. This index seeks to quantify the influence of geopolitical developments on global trade. The index uses open-source data across three verticals — diplomatic, military, and economic — to judge how things like trade interventions, conflicts, and diplomatic behaviour such as state visits rhetoric and voting patterns in bodies like the United Nations affect trade relationships. The latest reading in the index, for the year 2024, shows that trade risks were at the highest that year since 2019, the earliest year for which the Index has data.
