South Korea is prioritizing AI to boost its economy, with plans to invest $71.56 billion in areas like robotics and K-beauty starting in 2025, aiming to become a top-three AI power. However, OpenAI's CEO Sam Altman and an MIT study are warning about an AI bubble, as the study reveals that 95% of companies investing in AI haven't seen returns, despite tech giants like Meta, Microsoft, and Google planning to spend $320 billion on AI in 2025. Investors are also growing wary of special purpose vehicles (SPVs) used to invest in AI companies like OpenAI and Anthropic, citing high fees and potential scams. Meanwhile, experts predict AI bots will soon dominate the internet, requiring users to adopt their own AI tools for negotiation. Companies are actively developing new AI technologies; Tencent Cloud upgraded its AIoT product with voice control (TWeTalk) and video analysis (TWeSee), while Proton enhanced its Lumo AI assistant for better reasoning and privacy through encryption. To ensure ethical AI development, researchers are exploring the use of synthetic faces for AI training to protect personal data. Digital infrastructure is crucial for AI success, requiring investments in networks like 5G and data centers. In healthcare, Allegheny Health Network (AHN) doctors are using AI software from Abridge to automate paperwork, improving efficiency and patient care. Older adults are also starting to use AI tools, but have concerns about trust and privacy. Nvidia has launched new AI models for more languages, and Meta has added AI translations to Facebook and Instagram reels. Google has launched new Pixel phones with AI features and is expanding its AI Mode to more countries.
Key Takeaways
- South Korea plans to invest $71.56 billion in AI projects starting in 2025 to boost its economy and become a top-three AI power.
- OpenAI's Sam Altman and an MIT study warn of an AI bubble, with the study finding that 95% of AI investments haven't yielded returns.
- Tech companies like Meta, Microsoft, and Google plan to spend $320 billion on AI in 2025.
- Investors are concerned about risky SPVs used to invest in AI companies like OpenAI and Anthropic, citing high fees and potential scams.
- Experts predict AI bots will soon outnumber humans on the internet, necessitating AI tools for buyers to negotiate with seller bots.
- Tencent Cloud upgraded its AIoT product with TWeTalk for voice control and TWeSee for video analysis.
- Proton updated its Lumo AI assistant with improved reasoning and privacy through encryption.
- Researchers are exploring synthetic faces for AI training to enhance ethical AI development and protect personal data.
- Strong digital infrastructure, including 5G networks and data centers, is essential for AI success.
- Allegheny Health Network (AHN) doctors are using AI software from Abridge to automate paperwork and improve patient care.
South Korea prioritizes AI investment to boost economy
South Korea is making artificial intelligence a top priority to help its economy grow. President Lee Jae Myung's administration plans to introduce AI projects starting in 2025, investing in areas like robots, cars, and K-beauty. The government will create a $71.56 billion fund with the private sector to support these strategic sectors. They aim to become a top three AI power and increase the country's low growth rate. This plan also includes support for childcare, industrial safety, and digital assets.
South Korea prioritizes AI investment to boost economy
South Korea is making artificial intelligence a top priority to help its economy grow. President Lee Jae Myung's administration plans to introduce AI projects starting in 2025, investing in areas like robots, cars, and K-beauty. The government will create a $71.56 billion fund with the private sector to support these strategic sectors. They aim to become a top three AI power and increase the country's low growth rate. This plan also includes support for childcare, industrial safety, and digital assets.
South Korea prioritizes AI investment to boost economy
South Korea is making artificial intelligence a top priority to help its economy grow. President Lee Jae Myung's administration plans to introduce AI projects starting in 2025, investing in areas like robots, cars, and K-beauty. The government will create a $71.56 billion fund with the private sector to support these strategic sectors. They aim to become a top three AI power and increase the country's low growth rate. This plan also includes support for childcare, industrial safety, and digital assets.
OpenAI CEO warns investors about AI bubble
OpenAI CEO Sam Altman and other experts are warning about an AI bubble, where investors are too excited about AI. Tech companies like Meta, Microsoft, and Google planned to spend $320 billion on AI in 2025. However, an MIT study found that only 5% of businesses using AI have seen any profit. Altman says investors could get 'burnt' like during the dot-com bubble. He remains optimistic about future AI like GPT-6, but warns that the current high spending might not last if investors don't see results.
MIT study says most AI investments don't pay off
A new MIT study reveals that 95% of companies investing in artificial intelligence have not seen any returns. Researchers examined 300 AI ventures and found that despite $30-40 billion in investments, most companies didn't profit. This raises concerns about a possible bubble, similar to the dot-com crash. The study also suggests that buying AI tools is more effective than trying to build them in-house. Investors are now worried that the massive spending on AI may not be wise.
AI bots will soon dominate the internet
Experts predict that AI bots will soon outnumber humans on the internet. These bots will represent individuals and companies, interacting with each other to make decisions like pricing. AI-driven e-commerce will change prices based on who you are, updated by the microsecond. Buyers will need their own AI tools to negotiate with seller bots. This shift could lead to complex and wasteful interactions, making it harder to monitor what's happening online. Websites will need versions for both humans and AI agents.
Tencent Cloud upgrades AIoT for smarter devices
Tencent Cloud has updated its AIoT 2.0 product, adding advanced AI features to smart devices. The new version includes TWeTalk for voice control and TWeSee for video analysis. TWeTalk works with smart toys and headphones, offering natural conversations and WeChat integration. TWeSee improves security and industrial uses with object detection and language search. These upgrades aim to lower costs, improve efficiency, and support collaboration between devices and the cloud. Tencent Cloud is working with partners to use these solutions globally.
Proton's Lumo AI assistant gets smarter and more private
Proton has updated its Lumo AI assistant to version 1.1, making it faster and smarter while protecting user privacy. Lumo can now reason through complex problems 200% better and understands context 170% better. It also generates correct code 40% more often. Unlike other AI assistants, Lumo encrypts conversations so that Proton cannot read them, and it doesn't use personal chats to train the AI. Proton offers a paid Lumo Plus option for unlimited use, betting that people will pay for privacy.
Investors worry AI craze creates risky SPV bubble
Investors are concerned that the AI boom is creating a bubble with special purpose vehicles (SPVs). SPVs let investors pool money for deals in AI companies like OpenAI and Anthropic. Some SPVs have high fees and unclear structures, leading to scams. Experts like Bill Gurley warn against buying SPVs. OpenAI and Anthropic advise caution, saying some sales may not be valid. These concerns arise as AI valuations increase, and retail investors rush to invest in top AI startups.
AI needs strong digital infrastructure to succeed
Artificial Intelligence (AI) relies on digital infrastructure like networks and data centers to work. Strong infrastructure leads to better AI, while weak infrastructure limits access. Governments must invest in digital infrastructure to stay competitive in the digital economy. AI needs high-speed networks like 5G, powerful processors, and data centers to function efficiently. As AI models grow, they will need even more advanced infrastructure. Therefore, improving digital infrastructure is essential for developing AI.
Can fake faces make AI training more ethical?
AI has been known to discriminate against certain groups, especially in facial recognition. While accuracy has improved, it often comes at the cost of privacy, as companies use real faces to train AI. A new solution is using synthetic, or fake, faces to train AI algorithms. Although models trained on fake faces are less accurate than those trained on real faces, researchers are optimistic that this method will protect personal data while maintaining fairness. The goal is for everyone to have an equal chance of being correctly recognized.
AI news update: Investments, risks, and new tech
This week's AI news includes a study showing that most AI investments haven't paid off, raising concerns about an AI bubble. Experts warn that unchecked AI growth could cause job losses and environmental issues. Meta is restructuring its AI division, while Google launched new Pixel phones with AI features. Google is also expanding its AI Mode to more countries. A survey found that older adults are using AI tools, but have concerns about trust and privacy. Nvidia launched new AI models for more languages, and Meta added AI translations to Facebook and Instagram reels.
AI helps Allegheny Health Network doctors with paperwork
Doctors at Allegheny Health Network (AHN) are using AI to help with paperwork. They use AI software on their phones to record conversations with patients, with the patient's permission. The software then generates notes quickly, saving doctors time. AHN partnered with Abridge, a Pittsburgh-based company, to use this technology. The AI software is HIPAA-compliant, keeping patient data safe. Doctors report less burnout and patients feel more attentive because doctors spend less time taking notes.
Sources
- South Korea makes AI investment a top policy priority to support flagging growth
- South Korea makes AI investment a top policy priority to support flagging growth
- South Korea makes AI investment a top policy priority to support flagging growth
- OpenAI CEO joins chorus of industry experts warning about an "AI bubble"
- MIT study shows 95% of AI investments deliver no returns
- AI bots will soon overrun humans on the internet
- Tencent Cloud unveils AIoT 2.0 to integrate multimodal AI in global smart devices
- Proton’s privacy-first Lumo AI assistant gets a major upgrade
- SPVs in the hottest tech companies are suddenly everywhere. Investors are warning about a bubble popping as AI FOMO sweeps Silicon Valley.
- Digital Law: Artificial Intelligence: Digital Infrastructure
- Can fake faces make AI training more ethical?
- AI Update, August 22, 2025: AI News and Views From the Past Week
- Allegheny Health Network teams with Pittsburgh-based AI company to advance documentation
Comments
Please log in to post a comment.