Notes I took from the artificial intelligence news that caught my eye between January 19 and January 29.

  • Meta CEO and founder Mark Zuckerberg made statements about artificial intelligence on his Instagram account. He will reportedly utilize the processing power of 350,000 Nvidia H100s to develop ‘open-source’ (AGI, Human-level artificial intelligence) artificial intelligence. It is clear from the statement how excited he is. It is impossible not to be excited by future opportunities. I am sure that open-source super artificial intelligence is a lie, but I cannot prove it for now. I had previously seen Zuck this excited in the field of ‘Metaverse’ and ‘VR’, and the course of the Metaverse did not go very well. But the issue of artificial intelligence seems like it will change the rules of the game… To not see this, one truly needs to be ‘blind and deaf’.

The full text of the statement: “Some updates on our artificial intelligence efforts. Our long-term vision is to build general intelligence, open source it responsibly, and make it widely available so everyone can benefit. We’re bringing our two major AI research efforts (FAIR and GenAI) closer together to support this. We’re currently training our next-generation model Llama 3, and we’re building massive compute infrastructure to support our future roadmap, including 350k H100s by the end of this year – and overall almost 600k H100s equivalents of compute if you include other GPUs. We’re also really excited about our progress building new AI-centric computing devices like Ray Ban Meta smart glasses. Much more to come soon.”

https://www.instagram.com/reel/C2QARHJR1sZ

  • I discovered Aysu Dericioğlu’s channel on YouTube. She has taken notes on the panel attended by OpenAI CEO Sam Altman at the World Economic Forum currently being held in Davos. Since it is a super summary, there was no need to examine the artificial intelligence discussions that left their mark on Davos. Because it can be very exhausting.

  • Google’s artificial intelligence DeepMind’s algorithm named AlphaGeometry solved 25 out of 30 olympiad geometry problems within the standard olympiad time limit. This is above the number of problems an average person could solve. The development of AlphaGeometry represents a significant advancement in AI’s mathematical reasoning capabilities. This technology could open new paths for learning and researching geometry.
  • A team from the Indian Institute of Science in Bangalore, India, used artificial intelligence to identify the regions of the brain that control imagination and memory formation. Researchers found that the hippocampus and prefrontal cortex regions of the brain play an important role in imagination and memory formation. While the hippocampus helps with the learning and storage of new information, the prefrontal cortex helps with the retrieval of this information. Researchers developed a method to monitor the activity of these regions using artificial intelligence. This method helped researchers better understand the role of these regions in imagination and memory formation. Researchers say their findings could help develop new methods for the treatment of imagination and memory disorders. ( Source )
  • OpenAI is planning to add a Quick Settings shortcut to ChatGPT for Android. This will allow users to access ChatGPT directly from the Quick Settings panel without opening their home screen or the Settings app. If Apple and Google do not provide a solid response, in the medium term we might start saying ‘Hey GPT’ instead of ‘Hey Siri’ or ‘Ok Google’.
  • Google has brought three new generative AI features to Chrome. These features aim to help users be more efficient while browsing the web and performing productivity tasks.
    • The “Organize Tabs” feature helps users manage their tabs more efficiently. This feature allows you to automatically organize your tabs randomly or according to a specific order.
    • The “Create with AI” feature allows users to create custom themes for Chrome and helps you customize various elements such as background image, color scheme, and font.
    • The “Help me write” feature offers various tools that help you generate text, translate languages, and answer your questions while browsing the web.
  • Speaking of Google, here is another piece of news. The company announced Lumiere , its AI video generation system. Lumiere uses a space-time diffusion model, a new technique for generating realistic and high-quality videos. Lumiere can generate videos up to 5 seconds in length. The videos can contain various objects and environments. It is also worth adding that the artificial intelligence application uses real-world data to create the videos, making them look realistic. But of course, the decision is yours. Do you think it is realistic enough?
  • The US Federal Trade Commission (FTC) has finally stepped in regarding companies dealing with artificial intelligence. While the rise of artificial intelligence has just(!) begun, as the saying goes, ‘nip it in the bud’, it sent letters to Alphabet, Amazon, Anthropic, Microsoft, and OpenAI, asking the companies to explain the impact of their investments on the competitive landscape of generative artificial intelligence. In the inquiry launched to help artificial intelligence technology develop in a fair and competitive environment, the companies reportedly have 45 days to respond. ( Source )
  • Taylor Swift’s deepfake photos became a trending topic on the internet. These deepfake photos, which contained ‘inappropriate’ images, brought along a lot of discussions. Thanks to artificial intelligence, it is possible to make explicit images and even videos of anyone. I think the coming years will be the scene of these types of discussions. Therefore, there is plenty of time to discuss it here.