Yo, its Briefed. We’re like a Tootsie Lollipop - satisfying, delicious, & we don’t make you work too hard to get to the center (i.e. the good stuff.)

The Singularity hasn’t happened yet so in the meantime, enjoy some cool stories:

🦾 ChatGPT HUGE UPGRADE! Tree of Thoughts (Reasoning Improved 900%)

🤖 TikTok Is Testing an Artificial Intelligence Chatbot Named Tako

🦿 Artificial intelligence used in medical procedure to help paralyzed man walk

🪲 New superbug-killing antibiotic discovered using AI

🤖 ChatGPT HUGE UPGRADE! Tree of Thoughts (Reasoning Improved 900%)

Googles Deepmind came out with a new paper about Tree of Thoughts (ToT), which is basically like giving ChatGPT a map and compass to navigate its own mind. It's like the AI version of a self-help book, but instead of finding inner peace, it's finding better answers.

The paper starts off with a bang, claiming that this new method has boosted ChatGPT's problem-solving abilities from a measly 4% to a whopping 74%. That's like going from "I can't even solve a Rubik's cube" to "I can solve a Rubik's cube while juggling chainsaws".

The ToT method is all about allowing the AI to explore its own thoughts, backtrack, and make decisions. It's like giving the AI a backstage pass to its own thought concert. The researchers tested this method on a bunch of tasks, and it outperformed the old methods like a cheetah racing a tortoise.

One of the tasks was a game of 24, a mathematical challenge. The old methods scored around 4-9%, while ToT scored 74%. That's like comparing a kid's lemonade stand to a multinational corporation.

Then they tested it on a creative writing task, where the AI had to write four paragraphs ending with four given sentences. The AI generated several plans, voted on them, and then wrote a passage that was more coherent than most of my college essays.

Finally, they tested it on a 5x5 crossword puzzle. The old method scored a word success rate of below 16%, while ToT scored 60%. That's like comparing a paper airplane to a fighter jet.

In conclusion, the researchers found that by allowing ChatGPT to explore its own thoughts, they could significantly improve its problem-solving abilities. It's like they gave the AI a brain upgrade. So, what's next? AI stand-up comedians? AI novelists? Only time will tell. But for now, it seems like the AI world just got a whole lot more interesting.

🤖 TikTok Is Testing an Artificial Intelligence Chatbot Named Tako

TikTok is apparently testing a new AI chatbot called Tako. This isn't your average, run-of-the-mill chatbot. No, this one can recommend videos based on what you ask it. It's like having a personal assistant who knows your taste in cat videos and cringy dance challenges.

According to Daniel Buchuk, a guy who spends his time spotting these kinds of app changes, Tako could "radically change search and navigation" in the app. That's like saying adding wheels to a suitcase could radically change travel.

Now, Tako isn't just sitting there waiting for you to ask it something. No, it's proactive. It'll display suggested prompts to help you start a conversation. It's like that friend who always knows how to break the ice at parties.

TikTok's spokesperson, Zachary Kizer, called Tako "a limited experiment". It's currently only being tested in the Philippines. So, for those of us in North America or Europe, we'll just have to wait and see if we get to meet Tako.

Interestingly, TikTok recently filed a trademark application for "chatbot software" called Tako. That's like a big neon sign saying "Coming Soon". And they're not the only ones. Snapchat and Meta are also jumping on the chatbot bandwagon. It's like the social media world is having a chatbot party, and everyone's invited.

Well, it seems like TikTok is gearing up to make your scrolling experience even more personalized. And who knows, maybe Tako will be the one to finally understand your obscure video references.

AI GENERATED IMAGES OF THE DAY

celebrities as muppets / source

🦿 Artificial intelligence used in medical procedure to help paralyzed man walk

Here's a story that's going to make you believe in miracles, or at least in the power of technology. A man named Gert-Jan Oskam, who was paralyzed from the waist down after a motorcycle crash, is now walking again. And no, he didn't find a magic lamp with a genie inside. He had a little help from our friend, artificial intelligence.

After his accident in 2011, Oskam said he dreamt of walking again. He tried standing up, making steps, but it was like trying to win a marathon with a pair of flip flops. It just wasn't happening.

Fast forward to 2022, and Oskam is working with researchers at Lausanne University Hospital in Switzerland. They surgically inserted electronic implants in his brain and spinal cord, the areas that control movement. It's like they installed a new operating system in his body.

Then, with the help of AI, they built a "digital bridge" between his brain and spine, bypassing his injuries. It's like they created a highway for his thoughts to travel directly to his legs. Now, Oskam can think about moving, and his body follows his thoughts. It's like his body finally got the memo.

This is the first successful procedure of its kind. It's like they've just landed on the moon of medical breakthroughs. The researchers describe it as using AI as a thought-decoder, processing what the neurons in the brain region are trying to do, and sending that signal to the spine. It's like having a translator for your thoughts.

The technology is still in the preliminary stages, but the researchers' mission is to "bring it to other people." So, who knows? Maybe in the future, we'll all have AI-powered superpowers. But for now, let's just marvel at this incredible breakthrough.

🪲 New superbug-killing antibiotic discovered using AI

Scientists have turned to artificial intelligence (AI) to help them in their never-ending game of whack-a-mole with superbugs. And guess what? They've hit a mole. They've discovered a new antibiotic that can take down a particularly nasty species of superbug.

The AI was like a super-efficient personal assistant, sifting through thousands of potential chemicals to find the ones worth testing in the lab. The result? A potent, experimental antibiotic called abaucin. It's like finding a diamond in a pile of rocks.

Now, this isn't just any superbug. It's Acinetobacter baumannii, one of the three superbugs the World Health Organization has identified as a "critical" threat. It's like the Darth Vader of bacteria, often able to shrug off multiple antibiotics and causing havoc in hospitals and care homes.

The researchers trained the AI by feeding it information about the chemical features of drugs that could attack this problematic bacterium. It's like they gave the AI a crash course in chemical warfare. Then, they let the AI loose on a list of 6,680 compounds. In just an hour and a half, the AI produced a shortlist.

Out of 240 tested in the lab, they found nine potential antibiotics, one of which was the superhero antibiotic abaucin. It was so potent, it could treat infected wounds in mice and kill A. baumannii samples from patients. It's like they've found the Superman of antibiotics.

The next step is to perfect the drug in the lab and then perform clinical trials. The first AI antibiotics could take until 2030 to be available for prescription. It's like waiting for the next season of your favorite show.

Abaucin only works on A. baumannii. It's like a sniper taking out a specific target. This precision could make it harder for drug-resistance to emerge and could lead to fewer side-effects.

In conclusion, AI has proven to be a game-changer in the fight against superbugs. It's like we've just unlocked a new level in the game of antibiotic discovery. And who knows what other breakthroughs are just around the corner?

MEME OF THE DAY

1. See you beautiful cyborgs over on twitter @briefedai

2. Get your brand in front of thousands of AI Lovers by sponsoring Briefed.

3. Was this email forwarded to you? Signup Here

How Was Today’s Email?

• 💼💼💼💼💼The newsletter Gods would be proud

• 💼💼💼Not your best work

• 💼Worse than unnecessary work meetings

Send a reply to this email, we always respond :)