Photonic processors for faster, energy efficient AI
Researchers at MIT have developed a groundbreaking photonic chip that uses light to perform all key computations of deep neural networks, achieving faster and more energy-efficient processing than traditional electronic hardware. By integrating nonlinear optical function units (NOFUs) onto the chip, the system eliminates the need for external electronics, enabling ultra-fast computations in less than half a nanosecond while maintaining over 92% accuracy.
Fabricated using standard commercial processes, this scalable technology has the potential to revolutionize applications like telecommunications, lidar, and scientific research by significantly improving the speed and efficiency of deep learning systems.
Singularity 6 years away?
AI singularity, where AI surpasses human control, may be approaching as systems near human-level capabilities. Language translation, a complex challenge, serves as a key metric. Translated, a Rome-based company, tracked its AI’s progress from 2014 to 2022, showing translation accuracy steadily improving.
The time for editors to fix machine translations dropped from 3.5 seconds per word in 2015 to 2 seconds today, with human-level performance expected by decade’s end. This progress suggests AGI, capable of matching human skills across tasks, may soon become a reality, potentially transforming society.
Intel missed AI revolution
Intel's CEO, Pat Gelsinger, has stepped down as the company faces mounting challenges in the wake of the AI revolution. Once a leader in chip manufacturing, Intel has struggled to adapt to the growing demand for AI-capable processors. Unlike Intel’s traditional CPUs, Nvidia's GPUs excel in AI tasks due to their ability to handle complex parallel processing, making them the preferred choice for AI computing.
This has driven Nvidia’s market value to over $3 trillion, while Intel's has plummeted to $100 billion, with its stock declining 61% since Gelsinger took over. Despite efforts to regain competitiveness, including restructuring and government support through the CHIPS Act, Intel has lagged behind in meeting evolving technological demands, underscoring how the rise of AI-focused technology has drastically reshaped the chip industry, leaving legacy players like Intel at a disadvantage.
Musk vs. Altman
Elon Musk and his AI startup, xAI, have filed an injunction seeking to stop OpenAI from transitioning further into a profit-driven enterprise and allegedly preventing its investors from funding competitors, CNBC reported. The move is part of an ongoing feud between Musk, who co-founded OpenAI in 2015 but left in 2018 over concerns about its shift from a non-profit to a "capped" for-profit model.
Musk has accused OpenAI of abandoning its original mission of open AI research, while OpenAI has dismissed his legal actions as grievances stemming from his departure. OpenAI is now pursuing status as a for-profit public benefit corporation to attract more investment.
In case you’re interested — we’ve got hundreds of cool AI tools listed over at the Daily Zaps Tool Hub.
If you have any cool tools to share, feel free to submit them or get in touch with us by replying to this email.
