Buzz over Grok’s abilities

Reading time: 5 minutes  

 Newsletter | News | Tools 

Today we will discuss-

• 🌎U.S, U.K, others sign AI safety pact

• ⭐Musk's AI chatbot Grok creates stir, outshines OpenAI's ChatGPT

• 🦸‍♂️Alibaba’s AI model achieves breakthrough in early cancer detection

• 😱9 amazing AI tools you might not have heard of

All this and more - Let's dive in!

👩‍🍳What’s cooking in the newsroom?

U.S, U.K, 16 other countries sign agreement for safe, secure development of AI

♨️News - On 27 November, an international agreement on AI safety was signed by 22 law enforcement and intelligence agencies from 18 countries, including the U.S and the U.K. This aims to ensure that new AI products are designed with security in mind.

The agreement highlighted the importance for companies involved in designing and using AI to develop and deploy it in a manner that safeguards customers and the broader public from potential misuse.

🕵️‍♂️What issues does it address? While not legally binding, the agreement suggests general measures, including monitoring AI for misuse, securing data from manipulation, and evaluating software suppliers.

It also addresses concerns about how to keep AI technology safe from hackers and recommends practices such as releasing models only after conducting proper security tests.

However, it is important to note that the framework does not address complex issues surrounding the proper use of AI or the methods of data collection for AI data models.

🫣What's more? Surprisingly, AI powerhouse China was not part of the agreement. To this end, TSMC’s chairman Mark Liu, in a speech on the same day, said that the United States’ decision to exclude China could result in a worldwide innovation slowdown and contribute to the breakdown of global relations.

Elon Musk's AI chatbot Grok challenges ChatGPT’s abilities

☕News - Elon Musk’s AI chatbot Grok has made headlines over the last few days due to its remarkable abilities. Although details about Grok’s capabilities are still emerging, the chatbot seems to be gaining popularity as an entertaining source of information among users.

🤭What's been going on with Grok? 

• In a recent post shared on X, a user asked ‘Grok’ to roast Musk in a single word. The AI responded with humor, calling Musk ‘Overrated,’ drawing a cheeky acknowledgement from the Tesla boss who replied with, “Accurate.”

• Another post that recently went viral showcased a user using Grok to decode a tweet by Musk. The post captured the interest of many, even prompting a response from the tech billionaire.

• Joining the bandwagon, Musk himself posted a poem composed by Grok on the theme of love. What made it more interesting was that the untitled poem followed the traditional rules and guidelines expected of a poet when writing a sonnet.

This is impressive as in less than a month, Grok has proven to be wittier, smarter, and less prone to making mistakes than OpenAI’s ChatGPT. Especially since Grok is trained mainly on X’s data and user posts, while ChatGPT has been around longer and trained from various sources.

Alibaba’s AI model achieves 99.9% accuracy in early detection of ‘king cancer’

👨‍🔬News - Chinese scientists have developed an AI tool which can help in the early-stage screening of pancreatic cancer, commonly known as the "king of cancers."

😷Why is this important? 

• Pancreatic cancer typically has a survival rate of less than 10 percent over five years, mainly due to the difficulty in detecting it early.

• Moreover, according to the Mayo Clinic, pancreatic cancer is rarely found in its early stages when the chance of a cure is highest. This is so as it often shows no symptoms until spreading to other organs.

🤖How will the AI help? The AI model, developed by Alibaba Group’s DAMO Academy in collaboration with clinical researchers, has shown promising results in early screening for pancreatic cancer. The model boasts a 99.9% accuracy, indicating only one false-positive case per 1,000 tests, and a 92.9% sensitivity, surpassing radiologist performance by 34.1%. 

This is crucial as early detection significantly improves survival chances. Notably, high-risk patients identified early survive for an average of 9.8 years, while those diagnosed late only survive for 1.5 years.

It is worth noting that despite showing promising initial results, there is a long way to go before this technology can be applied in clinical practice. This, because AI-based imaging applications have not received approval from Chinese authorities yet.

👩🏼‍💻What else is happening?

• LG, UNESCO partner to promote AI ethics

• California seeks to prep workforce for coming AI boom

• Singapore's Changi airport tests AI security screening to cut time

• Consumers may face higher prices to protect employees from AI and automation, says EU Commissioner

• Chinese EV maker Nio aims to replace 30% of its workforce with AI, robots to improve efficiency, cut costs by 2027

🤓Interesting facts

• AI is getting smarter and could outsmart humans in the future. With its ability to learn, AI is expected to reach the intelligence level of adults by 2029, a significant advancement from its 2013 status, which was comparable to a 4-year-old's intelligence.

• The most commonly searched Google keyword is "chatGPT login," with nearly 6 million monthly searches.

• 83% of companies claim that using AI in their business strategies is a top priority. Among the most common applications are automated emails and chatbots, widely used in everyday business communications.

👩🏼‍🚒Discover mind-blowing AI tools

1. ClientReports.ai - An AI-powered platform that helps businesses generate client reports quickly and efficiently

2. Hey mind - An app that allows users to create mind notes, reflect on their thoughts, and engage in conversations with renowned thinkers

3. STLST - An app that offers personalized style advice, fashion ratings and feedback on outfits

4. Captioncue - An AI-powered tool that helps users generate captions that align with their unique style and voice

5. Venturekit - An AI-powered platform that simplifies the process of starting a business by providing valuable tools and resources to entrepreneurs

6. Elite-AI-Write - A content creation tool that helps users generate SEO-optimized and plagiarism-free content for blogs, ads, emails, and websites

7. Launch Teddy - An AI assistant designed to provide specialized support for product launches

8. GifStar - A tool that allows users to create and customize GIFs easily

9. MovieUncover - A tool that provides movie suggestions based on the user's preferences 

💥Mythbuster

Myth - AI algorithms can process any data

Fact - The myth that AI algorithms can process any data is simply not true. AI algorithms are trained on specific datasets, and they can only process data that is similar to the data they were trained on. For example, an AI algorithm that was trained on a dataset of text documents will not be able to process images or audio files.