Good morning! Grab your morning brew because here's your AI Trends dose, the newsletter that's like your trusted barista—always brewing the richest, strongest shots of AI, tech, and innovation updates.

Let's froth up today's news latte together:

First Quantum Computer With Over 1,000 Qubits Unveiled, A Tech Hit 💡🔢

Our future predictions are being surpassed by reality. This time it's Atom Computing, a startup not afraid to push established boundaries. They have recently announced that they have been conducting internal tests on a quantum computer with none other than 1,180 qubits. But, hold on, because there's more: they plan to make it available to the public next year.

Let me explain why this is so amazing. To begin with, this would be, if not the first, one of the first quantum computers with over 1,000 qubits, a significant advance considering that just two years ago, Atom Computing was working with a system of only 100 qubits. Moreover, although the error rate for individual qubit operations remains high, the company is confident that its technology can scale rapidly, offering fertile ground for work on quantum error correction.

🔬Given that this news is brimming with high-level science, remember: The only qubit that matters is the one that makes us smile!

This new system represents a huge step for Atom Computing and for the field of quantum computing as a whole. In fact, it essentially demonstrates that there is room to improve the effectiveness and efficiency of these dizzyingly complex machineries. Sounds impressive, right? And while it's important to keep in mind that the total qubit count does not guarantee superior quantum performance, this announcement underscores the incredibly rapid evolution in qubit capacity over a short period of time.

In conclusion, as the field of quantum computing continues to evolve at an astonishing pace, disrupting our usual perceptions of what's possible, Atom Computing keeps its focus on the big prize: Manufacturing a truly useful and practical quantum computer. Just as we eagerly anticipated Silicon Valley's advancements in the past, we now turn our sights to Atom Computing and its peers in anticipation. Long live the quantum wonders!

Hacker Challenge: USB with 235 Million in Bitcoin, But the Owner Refuses Help 🚫💰

Speaking of modern crypto fairy tales, the block wall separating Stefan Thomas from his 7,002 Bitcoins is only a password away. Specifically, a lost password that could open the encrypted USB, IronKey S200, where his crypto treasure keys reside. In an unusual twist of fate, a team of hackers called Unciphered boasts the ability to scale this wall.

And how do we know this? In a bold experiment, a journalist generated a random password, set it on the same USB model, and sent it to the hackers. The results were extraordinary and were achieved with astonishing speed. Thanks to a secret technique that bypasses the IronKey's setup which permanently erases its contents after 10 incorrect password attempts, Unciphered was able to decipher the password in a staggering total of 200 trillion attempts.

🔓 It seems that Stefan may have a second chance to attend his own Bitcoin dance.

But there's a problem. Stefan Thomas is reluctant to accept the help. In fact, he has already declined Unciphered's hackers' offer, as he is working with two teams that previously committed to helping him solve the situation. Neither of the two teams has been able to replicate the Seattle guys' trick, but Thomas seems willing to keep waiting.

The end of this cryptocurrency story is yet to be written. We may never know if Stefan Thomas's lost bitcoins will be recovered or if they will stay permanently stuck in digital limbo. What we can conclude is that in the strange world of cryptocurrencies, even a forgotten password can become the beginning of an intriguing adventure.

'Eureka': GPT-4 Driven Algorithm Promises to Accelerate Robot Learning 🤖⚡

In a quantum leap for artificial intelligence and robotics, a team of scientists from Nvidia, the University of Pennsylvania, Caltech, and the University of Texas in Austin have revealed an astonishing algorithm. Named Eureka, this algorithm uses OpenAI's GPT-4 language model to design "reward functions" that promise to dramatically accelerate the learning of robots.

Eureka employs a reinforcement learning approach to enhance robot dexterity and bridge the gap between high-level reasoning and low-level motor control. With a hybrid gradient architecture, this algorithm can learn complex tasks of navigating through massively parallel simulations. According to the researchers, Eureka outperforms reward functions written by humans, signifying a revolution in the way we train robots.

🤖 If Skynet is taking notes, this might be the right time to start getting on good terms with our future robotic overlords!

Advancements from this algorithm can have massive applications in the interactions of robots with the real world. With features such as 'human feedback' and simulations in computer game-like worlds, it is possible that in the future we can teach robots new tricks at an accelerated pace. However, all this will be driven by AI models against a backdrop of simultaneous trials in many virtual dimensions.

Lastly, the implementation of Eureka is a clear example of the path we are taking towards merging AI and robotics. What's most exciting is that this is just the beginning. The team behind Eureka has made its research and codebase available for other scientists and robotics enthusiasts to continue refining and building upon this breakthrough. So, who knows, next time it might not be 'Eureka', but 'Voila' with GPT-5.

AI Bites 🍪

Xpeng starts removing HD maps from Tesla FSD-like feature in China. Xpeng is often called the Chinese challenger to Tesla for its efforts to bring advanced driving capabilities to its electric vehicles.

UMG files landmark lawsuit against AI developer Anthropic. Universal Music Group (UMG) has filed a lawsuit against Anthropic, the developer of Claude AI.

Amazon’s new generative AI tool lets advertisers enhance product images. Amazon is rolling out a new AI image generation tool for advertisers to generate backgrounds based on product descriptions and themes.

Watch this, thank me later... 👀

🥽🖥️ The virtual monitor functionality in the Meta Quest 3 augmented reality glasses promises to revolutionize the way we interact with the digital and physical world simultaneously 👇

canal.mutuo

View more on Instagram