AI, ML, LLM & Related Areas In The World : News &:Discussions.

[trombone]

as if deepfakes were not increasingly hard to detect, tiktok bytedance releases OmniHuman-1 which now makes it even harder and it's still not pubic but in lab!

ByteDance's OmniHuman-1 shows just how realistic AI-generated deepfakes are getting

ByteDance demoed a model that its researchers say creates realistic full-body deepfakes from a single image.

www.yahoo.com

View: https://www.youtube.com/watch?v=n6HKcS2pj0Q&t=18s

 

[Hari Sud]

Action Summit on Rapid AI Developments in Paris

Over the past two years, artificial intelligence (AI) has advanced at an unprecedented pace, with the U.S. leading the way. However, just four weeks ago, China disrupted the landscape by introducing a new AI system that rivals Open’s expensive technology at a fraction of the cost.

For a developing country like India, which is at the forefront of computer technology and building its own AI systems, these developments are crucial. AI holds immense potential to transform sectors such as healthcare, education, and defence while also making government operations more efficient. Moreover, Indian talent plays a significant role in high-tech innovation, particularly in Silicon Valley. As AI evolves, experts estimate its impact on trade could drive a 14% increase in global commerce.

Prime Minister Narendra Modi is in Paris to co-chair the Open AI Action Summit on February 10-11 alongside French President Emmanuel Macron. The summit aims to ensure that AI development remains a global effort rather than an exclusive U.S.-China competition. Leading tech executives from around the world are also attending.

A key focus of the summit will be establishing frameworks to regulate AI in a safe, ethical, and trustworthy manner. As this technology rapidly evolves, responsible oversight will be critical to maximizing its benefits while mitigating risks.

 

[vicky]

as if deepfakes were not increasingly hard to detect, tiktok bytedance releases OmniHuman-1 which now makes it even harder and it's still not pubic but in lab!

ByteDance's OmniHuman-1 shows just how realistic AI-generated deepfakes are getting

ByteDance demoed a model that its researchers say creates realistic full-body deepfakes from a single image.

www.yahoo.com

View: https://www.youtube.com/watch?v=n6HKcS2pj0Q&t=18s

A moment of silence for those who dream of becoming actors, YouTubers, Instagram influencers, or "Cornstars"



View: https://youtube.com/shorts/9CrP5gF8TKo?si=HR3yNmd0lRQd4suu

 

[Hari Sud]

Why the US and UK Refused to Sign the International AI Declaration

At the AI Global Summit, sixty countries signed the declaration that emerging AI technology remains open, inclusive, and ethical. However, the US and UK declined to sign the final declaration. The US cited concerns over excessive regulations despite the declaration not being regulatory in nature while the UK pointed to national security issues, which the declaration was not intended to address.

Major sponsoring nations such as France, India, and China fully endorsed the declaration, calling it a sustainable framework for both people and the planet, with immense potential.

The US has led the development of AI, transforming Open AI from a mere concept into a scientifically sustainable field with applications in science, medicine, commerce, and daily life. This path required significant intellectual investment and massive financial resources to develop advanced chips, GPUs, and AI data centres across the country, an undertaking only the US could afford at such scale.

Meanwhile, China has achieved similar advancements with far fewer resources and at a fraction of the cost. This disparity has not sat well with US policymakers, who are already wary of intellectual property theft and competition. The US remains committed to its high-cost, high-investment approach and is reluctant to share its innovations, particularly given concerns that its breakthroughs may have been copied or repurposed by China.

As for the UK, it remains unclear why national security concerns prevented it from signing the declaration.


Moving forward, summit organizers must engage with both the US and UK to address their concerns and encourage broader international cooperation in AI governance.

 

[Hari Sud]

OpenAI and DeepSeek: The AI Battle

Like it or not, DeepSeek has shaken up the dominance of well-established players like OpenAI and Gemini. By offering open-source licenses at low costs, DeepSeek is aggressively attracting applications that need access to its AI system.

This follows a familiar pattern seen over the last 25 years as China has consistently reverse-engineered or replicated products and sold them in Western markets at a fraction of the cost. Now, the same strategy is playing out with DeepSeek. While its lower pricing is a major draw, concerns remain about the quality of information it provides. Still, many users will be drawn in by the affordability.

The potential applications of artificial intelligence are limitless. At its core, AI is the replication of human cognitive processes through computer algorithms once a concept of science fiction, now a reality in its early stages, poised for rapid evolution.

The U.S. is making bold moves, with President Trump spearheading a $500 billion initiative to accelerate AI advancements. During his visit, Prime Minister Modi sought collaboration on this ambitious project. However, India’s private sector has shown limited enthusiasm, leaving government funding as the only viable path forward. Falling behind in AI development is not an option, it is the future.

So far, the only major AI investment in India has come from Reliance and Mukesh Ambani, who plan to build one of the world’s largest data centre to support local AI development. India’s Silicon Valley must act swiftly, or it risks becoming permanently dependent on imported AI tools, an unfavourable scenario for the world’s leading software supplier.

 

[Hari Sud]

Is AI a Transformative science?

Yes, it is, in fact it is game changing tool which will change the way we live work & communicate. As a matter of fact, AI acts as a librarian, retrieving hidden knowledge rather than generating original thought. A noted commentator pointed out that AI doesn’t just streamline processes—it transforms access to information, unlocking opportunities for millions.

OpenAI pioneered this revolution with general-purpose language models powered by high-end chips and massive data centres—a costly approach. But just a month ago, DeepSeek achieved similar results while excelling in logical reasoning and mathematics, often surpassing OpenAI in complex problem-solving.

DeepSeek’s founder, a former hedge fund manager, has shaken the U.S. AI landscape, wiping out trillions in market value from top AI and chip firms. This also raises concerns about China replicating Silicon Valley innovations.

Security architecture is a key differentiator. OpenAI enforces robust security and strict model control, whereas DeepSeek’s open-source framework allows greater customization but demands more user vigilance. Security, along with cost, remains OpenAI’s key advantage.

DeepSeek is now a major competitor, alongside Grok (52% market share), Optimole (11%), and Drift (9%). OpenAI still dominates, with 54% of its users in the U.S., followed by India and the U.K. (both at 8%).

Notably, India, a major AI adopter. It is leveraging technology to transition from an agricultural to a digital economy—QR code-based transactions are already a daily norm. Sam Altman, the Open AI, CEO has realized its importance for India, has paid visits to India and talked to government and industry officials.

Today we see users not fully aware of the AI world. It is used in search engines like Google, digital assistant like Alexa, social media… etc.