Big moves this week: @Meta launched new rayban smart glasses with built-in AI.But a critical question remains → How do we trust AI wearables with sensitive data and daily interactions? https://t.co/uEPGjKUSzK ...

Join @Tony0kAI alongside @Tiancheng_Xie @jiahengzhang96 and @sampearton for a week of live shows and fun.Verified community members engaging will win USDT. Make sure to join our Discord to be eligible for rewardshttps://t.co/uGLwFzaD5p https://t.co/y4ekjoial1 ...

AI just leapt out of the screen:– OpenAI building its own device– Meta’s glasses read your muscle signals– Harvard reversing disease cells– Albania appoints an AI as a ministerBut if AI runs our bodies & governments—how do we trust it?That’s why we’re building zkML, zkGPT, and Polyhedra ID. ...

Event Overview - THE EGGYS Season 1 is live, aiming to discover the next viral Meme Coin [1] - Competition period: September 19th to October 5th, winners announced on October 9th [1] Participation Guidelines - Launch a Meme Coin on EGG Coin Market [1] - Create a 5-second AI video using tools like PIKA, Luma, or Kling [1] - Showcase the Meme Coin on EGG Explore [1] Rewards - Top 100 participants win prizes [1] - Raffles for voters [1]

2/Proof Aggregation in Multi-GPU setupProof Aggregation in MPI Environment: Began integrating proof aggregation functionality under a multi-GPU MPI environment. This aggregation not only coordinates distributed provers and combines partial results seamlessly, but also reduces the overall proof size, improving efficiency in both storage and verification.Stay tuned for more updates 🔥 ...

1/Multi-GPU AccelerationPolynomial Evaluation with NCCL: Completed multi-GPU acceleration for polynomial evaluation using NCCL. The implementation enables efficient parallel computation across multiple GPUs, significantly improving throughput and scaling performance for large proof workloads. ...

Product Development - This week the company delivered key improvements across ZKML and Expander [1] Technology Focus - The company's focus areas include ZKML (Zero-Knowledge Machine Learning) and Expander [1]

5/Officially deprecated the GKR Square component. Removed redundant parts in the scratchpad that were tied to it, simplifying the codebase and reducing overall computational overhead.Stay tuned for more updates 🔥 ...

4/Introduced support for hints in the computation graph definition. This improvement reduces the number of witnesses required, streamlining execution and improving performance. ...

技术进展 - 团队在本周在 ZKML 和 Expander 方面取得了重大更新 [1]