Notes I took from the artificial intelligence news that caught my eye between May 4 and May 10. (Due to the intensity of my other work, I am taking an indefinite break from AI news notes.)

  • Warren Buffett, known for his investments in the stock markets, spoke about artificial intelligence at Berkshire Hathaway’s annual meeting on Saturday. He reportedly thinks artificial intelligence has great potential for good and bad. So much so that he said AI is comparable to nuclear weapons and, as far as I understand, its potential for scamming is higher. He also reportedly said he has some concerns about the control of AI. As the countdown for the arrival of AGI (artificial general intelligence) has begun, I think it is difficult for everyone to predict what the future holds.
  • According to research by the research firm Sprout Social for its 2024 Marketing Report on the social content consumption experiences of 2,000 participants in the US and the UK, the vast majority of Generation Z showed that they do not care whether social media influencers are real people or not.
  • YouTube has started testing a new feature for some Premium members. You can see if it is available for you to try the AI-powered fast-forward feature at https://www.youtube.com/new . With this feature, it is reportedly possible to automatically skip the parts that most users skip.
  • Generative artificial intelligence looks like it will design new drugs on its own in the near future. Scientists at Eli Lilly were literally amazed by the novel design of the molecules that artificial intelligence produced as part of drug discovery research. Experts at Lilly and Nvidia are reportedly saying that within a few years, artificial intelligence will produce not only new drugs but also drugs that humans could not create.
  • Rumors continue about Apple’s artificial intelligence adventure, which has turned into a never-ending story. According to the latest rumor, it is said that there will be ‘Photoshop’-level AI-powered editing tools in Apple Photos. The move that Apple, which could not even use the letters ‘AI’ more than a few times at the iPad launch event held on May 7, will make is eagerly awaited by everyone in the industry. I feel like if its move against its competitors falls weak this year, it might be too late.
  • Can OpenAI rival ‘Google Search’ without looking at its own size? Microsoft might perhaps realize the dream it has had for years through OpenAI. According to the news published in Bloomberg , OpenAI says it will start sharing the links of the answers given and even add visuals to the responses. Yes, this is not exactly like a search engine, but wasn’t the search engine of the future going to be like this anyway? Reaching the direct answer without even having to go to the sites, in the format and length we want. update According to the new rumor, the search.chatgpt.com address is said to be launched one day before the May 14 Google IO event. There are only 3 days left. We will see…
  • According to a news report on Wired.com , it was written that OpenAI is exploring how AI-generated porn could be used responsibly. Not for myself, but for a friend 🙂 I skimmed the content of the news a bit. OpenAI published draft documents on Wednesday indicating how it wants ChatGPT and other AI technology to behave. A part of the Model Specification document revealed that the company was exploring a transition into porn and other explicit content. Although this research is somewhat to get ideas from users, there are very thorny issues regarding the future of artificial intelligence in the porn industry. At the same time, it contains greater potentials than ever before. It’s a double-edged sword. Experts are highly concerned that AI-generated pornography could be used for harassment and exploitation. Attention curious friends.
  • Google DeepMind and Isomorphic Labs introduced AlphaFold 3, a new artificial intelligence model for the development of drugs and treatments. AlphaFold 3 is unrivaled in predicting the complex structures and interactions of life’s fundamental molecules, such as proteins, DNA, and RNA. This feature plays a key role in developing new drugs.
  • TikTok will reportedly start marking AI-generated content with a watermark to help users understand what is real and what is not. TikTok reportedly says its efforts are a step toward preventing the spread of misinformation on the social media platform.
  • Toyota announced that it will invest approximately $13 billion (2 trillion yen) in electric vehicles and artificial intelligence next year. These investments will reportedly focus on strengthening supply chain management and developing new forms of mobility, including electric and hydrogen vehicles. Toyota will also reportedly increase its efforts in adopting artificial intelligence for applications such as autonomous driving.
  • When talking about artificial intelligence and robotics technologies, the subject undoubtedly comes to weapons as well. The US Marine Corps has started testing AI-powered and armed robot dogs. I don’t believe that what is shown to us is even one percent of what this technology possesses, but I think they had to start getting people used to it from somewhere. The example in the video seemed to me to be at a level that even high school students could do today.