La artificial intelligence revolution it has, in fact, already begun. Not a day goes by without news about this new technology that is literally changing the way we live. As said by Alessandro Fugnoli, Kairos strategist, in his recent podcast "Al 4° piano", AI is “a revolution, more than the Internet or the railways. In ten years' time, 2023 will probably be remembered for being the year in which Artificial Intelligence, after seventy years of incubation, sharply accelerated its course and in which the markets began to take note of it".
In these hours, two announcements have aroused particular interest: the new sNvidia supercomputer based on artificial intelligence and benefits that this technology can bring in the medical field by creating an antibiotic, in just two hours, to fight off a powerful superbug.
Nvidia: AI-powered supercomputer ushers in a new era of computing
Nvidia, an industry-leading manufacturer of graphics processors, unveiled a artificial intelligence supercomputer (AI) called DGX GH200. The supercomputer will assist tech companies in building similar generative AI models like the OpenAI company's ChatGPT. The announcement comes just days after the American chip giant has revealed predictions of rapid sales growth, +50% revenues and profits in the quarter and the prospect of growing the same in the next for a capitalization of 960 billion dollars. The president and CEO of Nvidia, Jensen Huang, defined this platform as a “turning point for a new computing age“. Google Cloud, Meta and Microsoft will be the first companies to have access to this innovative platform, which is designed for "exploratory research in a pioneering way," says the company's CEO.
The new supercomputer is based on an sseries of 256 Nvidia superchips belonging to the family Gh200 Grace Hopper, which work together with Nvidia's NVlink Switch System technology. This allows you to create a GPU (graphics processing unit) unified of enormous dimensions, capable of offering performance of 1 Exaflop and a shared memory of 144 Terabytes.
And Artificial Intelligence discovers an antibiotic in two hours
Thanks to 'use of Artificial Intelligence, a research team led by McMaster University and the Massachusetts Institute of Technology has managed to discover a new and powerful antibiotic in just two hours. This antibiotic is effective against the superbug Acinetobacter baumannii, one of the most dangerous antibiotic-resistant bacteria in the world, according to the World Health Organization (WHO). The AI algorithm used in the trial could revolutionize the field of antibiotic discovery, speeding up the search for effective treatments for many other multi-resistant bacteria.
To train the algorithm, the researchers lab-grown the bacterium and got it exposed to about 7.500 different chemical compounds. The structure of each molecule was then fed to the algorithm along with information about their effectiveness. L'AI examined a library of 6.680 molecules never evaluated before and has selected the 240 most promising killer candidates in just two hours.
Next, the researchers tested the selected compounds in the laboratory and identified the new antibiotic specific for the A. baumannii bacterium. This feature prevents the bacterium from rapidly developing resistance to the drug.
The success of this research paves the way for the possibility of significantly accelerate the process of discovering new antibiotics to combat a wide range of microorganisms that threaten human health and beyond. The use of Artificial Intelligence could play a fundamental role in the fight against antibiotic resistance and improve the ability to deal with the most dangerous bacterial infections.
The moves of the USA and Europe to regulate AI
The dangers of one comes from artificial intelligenceand, however, still remain to be clarified. The same creator of ChatGPT, Sam altman (CEO of OpenAi), during a hearing in the US Senate, had launched a new alarm on artificial intelligence which “if left free it can hurt the world”. If it is true that its use, especially in the medical and research fields, can bring great benefits and it is also true that, without rules and an agency that limits the application, the risks for the correct management of democracy, elections and social conflicts grow which can be influenced or even distorted by false communications passed off as true. Europe has also moved to reach an agreement on the first law on artificial intelligence. This proposal, known as AI Act, is a regulation which applies uniformly in all Member States of the European Union and contains 85 articles and annexes. It is the first legislative text that aims to comprehensively regulate the field of AI. The text is under discussion but even if it is approved, the law may not come into force before 2025.
And there is also the proposal of an immediate voluntary pact
Time too long? In the meantime, to shorten the time, the European Commission is launching the proposal for a voluntary pact on artificial intelligence. Recipients of Big Tech. The proposal was discussed at the G7 on digital in Takakasi and was subsequently addressed during the meetings between the EU commissioners and the Google CEO, Sundar Pichathe. “We need the EU law on artificial intelligence as soon as possible. But the technology evolves at extreme speed. So we need a voluntary agreement on universal rules for AI. There is no time to lose," said the Vice-President of the European Commission, Margrethe Vestager. The initiative will be discussed today at the EU-US Trade and Technology Council in Lulea, Sweden.
Without proper AI regulation, caution must always be exercised
On the dangers that can arise from the use of artificial intelligence, we show a video by Morgan Freeman which invites to question reality. The problem is that the speaker is not Morgan Freeman at all, but he is a perfectly believable deepfake. The effect is incredibly realistic, difficult to identify if he weren't announcing himself as a fake “I'm not Morgan Freeman and what you see is not real, at least not in contemporary terms. What if I told you I'm not even a human being? Would you believe me?”. All video and audio is generated by artificial intelligence. And with the US elections of 2024 upon us, the risk that this technology could compromise democracy is just around the corner.
