Hot AI News of the Day (Last 24 Hours): Market Moves, Startups & Developments

Artificial intelligence is still riding the hype bubble, and right now the industry news is dominated by headline-grabbing product announcements, corporate name-dropping, and alarming predictions by artificial general intelligence (AGI) doomsayers—let alone AGI development plans and promises of a better future. From billion-dollar valuations to tools that bring AI within reach, here’s today’s roundup of the news you shouldn’t miss.

Startup Moves & Funding

Among the most notable today is Indian startup Blinkit-AI, which has secured USD 1.2 million in new funding. The company says it will use the funding to embed AI functionality across its logistics, digital commerce, and customer experience platforms. This is another example of India positioning itself as an increasingly prominent AI innovation hub.

Industry pundits have expressed concern that successful AI firms are currently caught in a valuation bubble. Global investment trends indicate that AI-driven companies grabbed 58% of VC investments in the first quarter of 2025, with some early startups seeing valuations between USD 400 million and USD 1.2 billion per employee. Comparisons are being drawn to the dot-com era, where expansion outpaced sustainable business fundamentals.

Another notable development is Gradient Ventures, a Google-funded VC firm focused on AI, which is officially spinning out as an independent firm under the new name Grdnt LLC. Freed from Google’s oversight, the company will take its next step as an independent investor with a new USD 200 million fund, an attractive prospect for founders who might otherwise hesitate to accept strategic funding from potential competitors.

Adding to the momentum, Thinking Machines Lab, a stealthy AI company co-founded by former OpenAI executive Mira Murati, has announced its first public tool: Tinker. The platform promises to make it easier to fine-tune large language models while abstracting much of the complexity around distributed training. Developers can now train open-source models like Llama or Qwen with relatively low overhead, significantly lowering the barrier to AI customization.

Summary:

• Blinkit-AI: $1.2M funding, expanding AI capabilities in logistics and commerce.
• Gradient Ventures → Grdnt LLC: Independent VC with $200M fund.
• Thinking Machines Lab → Tinker: Simplified fine-tuning for large AI models.

Between the funding and product news, these developments reflect the early, high-energy phase of AI growth: abundant capital, increasingly accessible tools, and ongoing concerns about overvaluation.

Infrastructure & Hardware Breakthroughs

On the hardware side, Euclyd, a European AI infrastructure startup, unveiled a new system called CRAFTWERK, featuring:

• 16,384 custom SIMD cores
• 1 terabyte of ultra-bandwidth memory per module
• Throughput of up to 8 petabytes per second
• System-level compute exceeding 32 peak petaflops at FP4 precision
• Rack-scale potential of more than 1 exaflop

If successful, CRAFTWERK could directly compete with incumbents like Nvidia and AMD, offering efficiency gains for massive AI model training. This demonstrates the escalating infrastructure arms race in AI, where performance and energy efficiency are key differentiators.

Additionally, Perplexity AI announced that its Comet browser, previously priced at $200 per month, will now be offered free of charge. The AI browser aims to combat low-quality online content with real-time precision information retrieval. A lower-cost subscription at USD 5 per month will provide premium publisher content and a new revenue-sharing model with media houses.

Google continues to support startups by offering technical guides and video tutorials for scaling AI agents. Topics include:

• Retrieval-augmented generation (RAG)
• Agent-to-agent orchestration
• API integrations

These resources are designed to help startups build AI-based products on Google Cloud.

Products, Platforms, and Agent Trends

In the past 24 hours, new products have further advanced the frontier of “agentic AI.”

• Tinker: Transitioning AI development from manually-intensive training to user-friendly, abstracted platforms.
• OpenAI’s Sora (v2): A video generator app in pilot across North America, allowing users to:
  • Create vertical videos from text prompts
  • Remix personal video clips with AI-generated elements

Sora’s release positions OpenAI in competition with platforms like TikTok and YouTube. The app includes safeguards such as watermarks and AI-content labeling to distinguish synthetic content from authentic media.

AI Agents:
These systems go beyond traditional chatbots, performing tasks such as:

• Reasoning
• Fetching data from external sources (APIs)
• Executing multi-step workflows

This development signals a shift from reactive AI assistants to proactive workmates, transforming both business and consumer use cases.

Trends & Risks Emerging

1. Valuation Bubble Concerns
   • High valuations may outpace revenue growth, risking market corrections.
2. Push Toward Openness
   • Tools like Tinker and open-source fine-tuning frameworks indicate a move away from proprietary AI models.
3. Infrastructure Race
   • Hardware breakthroughs are increasingly important as models grow in size and computational demand.
4. Rise of AI Agents
   • Multi-step autonomous AI workflows could reshape productivity but raise reliability and ethical concerns.
5. Content Authenticity & Regulation
   • Labeling, watermarking, and copyright protections are essential as AI-generated media proliferates.
6. Global Diversification
   • While the U.S. leads in investment, India and Europe are emerging as significant AI innovation hubs.

What to Watch Next

• Will AI startup valuations stabilize or continue inflating?
• Can infrastructure challengers compete with incumbents like Nvidia?
• To what extent will tools like Tinker empower small teams to innovate?
• Will OpenAI’s video initiatives reshape social media trends?
• How will regulators respond as AI-generated content expands across industries?

Conclusion

The last 24 hours illustrate both opportunity and risk in the AI boom. Startups are securing capital, industry leaders are returning with ambitious tools, and infrastructure innovators are challenging incumbents. Meanwhile, investors question sustainability, and consumers confront the implications of AI-generated content at scale.

The AI sector is entering a new phase marked by:

• Rapid democratization of technology
• Heightened hardware competition
• Increased scrutiny on ethics and economics

Ultimately, innovation and sustainability will determine which players become long-term leaders and which remain caught in the hype cycle. As today shows, in AI, a single day can dramatically reshape the conversation about technology’s future.