How AI is using music to read minds...

Good morning, and happy FRY-day! Grab your favorite dipping sauce, and let’s munch on the latest news from the world of AI. 🍟

🔨 AI TOOLS OF THE DAY

🙋‍♀️ WTF Does This Company Do?: This tool explains a landing page using just the domain. → check it out

👨‍🏫 Heights AI: Build an online course in less than a minute! Generate recommendations, a lesson outline and a description, plus even choose a relevant cover image—all you have to provide is the course title. → check it out

🎙️ Wize: A tool that taps into the wealth of knowledge shared in popular podcasts. check it out

ANOTHER CHIP MANUFACTURER TRIES TO CATCH NVIDIA 💾

AI chips are in high demand, and Nvidia—the leading chip manufacturer—better watch its back! 👀

What’s up? Tenstorrent, an AI chip developer, declared the completion of a $100-million funding round yesterday. The funding was led by major manufacturing companies Samsung and Hyundai, along with other investors such as Kia, Fidelity Ventures, Eclipse Ventures, and Maverick Capital.

What will the money be used for? The funds raised in this round will be utilized to accelerate Tenstorrent's product development and advance its AI chiplets' design and development. This will also help them progress on their machine learning software roadmap.

AMD TO WORK AROUND SANCTIONS, CREATE AI CHIP FOR CHINA 🇨🇳

China is begging for AI chips, and AMD is ready to deliver! 💾

What’s new? AMD, the semiconductor giant, has announced plans to develop an AI chip specifically for the Chinese market. This move comes as a response to the U.S. export curbs, aiming to comply with the restrictions while tapping into the potential of the Chinese market. Lisa Su, CEO of AMD, emphasized China's importance as a market and stated the company's commitment to being fully compliant with U.S. export controls.

How will this work? AMD's plan involves creating an accelerator chip, a type of semiconductor crucial for training large datasets in AI applications. The company aims to offer AI solutions to the Chinese customer base seeking such technology. To compete with market leaders like Nvidia, AMD is increasing production of its MI300 chip, positioning it as a rival to Nvidia's graphics processing units (GPUs) used for AI training. This strategic move aligns with AMD's goal of growing its data center business, with the MI300 chip expected to drive around 50% growth in the second half of the year.

What does this mean? AMD's decision to develop a specialized AI chip for the Chinese market highlights the significance of China as a lucrative market for U.S. chipmakers, especially in the AI sector. With Nvidia and Intel already pursuing similar strategies, AMD aims to challenge Nvidia's dominance in AI-related graphics processing. By complying with U.S. export controls while catering to the growing demand for AI solutions in China, AMD hopes to bolster its presence and seize a share of this expanding market.

AI USES MUSIC TO READ MINDS 🧠

AI is now reading minds, through music! 🎵

What’s up? A new study shows that AI can create a song that aligns with the genre, rhythm, mood, and instrumentation of music a person has recently listened to by analyzing their brain activity.

How was the study conducted? The process of the study involved two stages: training the Brain2Music AI to identify links between brain activity and musical elements, and then using the MusicLM AI to generate music based on the identified patterns. Brain scans from five participants listening to various music genres were used to create personalized AI models. The AI's ability to accurately capture the original music's mood and instrumentation was around 60%, with classical music being the most distinguishable genre.

How does this work? Using brain imaging data collected via functional magnetic resonance imaging (fMRI), the researchers trained an AI program to find links between musical features (such as genre, rhythm, mood, and instrumentation) and participants' brain signals. The AI was customized for each individual, identifying unique brain activity patterns associated with musical elements. Once trained, the AI converted previously unseen brain imaging data into musical elements of original song clips. This information was then fed into another AI model called MusicLM (developed by Google) to generate musical clips resembling the original snippets.

What does this mean? The development of Brain2Music opens up new possibilities for understanding how the brain processes music. By correlating brain activity with musical elements, researchers gain insights into which brain regions are involved in processing music and its meaning. The research could also pave the way for reconstructing music that people imagine in their minds, not just music they listen to. Moreover, this technology may have implications for individuals who have difficulty expressing themselves through speech, as it offers a potential avenue for translating their thoughts and emotions into music. 🎶

FRY-AI FANATIC OF THE WEEK 🍟

Congrats to our subscriber, bobloder! 🎉

Bobloder gave us a fresh and crispy and commented, “Great at keeping me up to date on AI solutions.”

(Leave a comment for us in any newsletter and you could be featured next week!)

HAS AI REACHED SINGULARITY? CHECK OUT THE FRY METER BELOW