© IE Online Media Services Pvt Ltd
Tags:
Journalism of Courage
Artificial Intelligence is moving at a brisk pace with many billionaires investing in companies that are accelerating their research and development in AI products and services. Even as the clamour seeking regulations gains momentum, there have been numerous AI experts who have time and again urged the world to take a step back and reflect on the changes that AI can affect.
Renowned author and documentary filmmaker, James Barrat, who was at the Digital Bridge 2023, held in Astana, Kazakhstan, delivered a captivating speech that painted a stark image of the challenges posed by Generative AI systems and Large Language Models (LLM)created by tech behemoths such as Google, Meta, OpenAI, and Microsoft.
Barrat in his evocative speech said that intelligence, not strength or speed, is the superpower that enables humans to shape the future. “Yet, are we unwittingly locked in an unwinnable intelligence race with the very future of humanity hanging in the balance?” he asked the audience.
The Generative AI systems and LLMs developed by the tech giants are trained on vast sets of content and texts from the Internet. According to Barrat, the ongoing advancements in AI are reflective of both our technological capabilities as well as the ethical quagmires they lead up to. In his speech, Barrat raised a critical question: Are these systems, which generate words and images, a boon or a bane?
The 63-year-old warned that while Generative AI offers monumental profits to a select few businesses and benefits to our lives, it also challenges our privacy, jobs, and even our society. He believes that the core of the matter is the inadequate compensation for creators of the content that these AI systems are trained on. He added that there are numerous copyright lawsuits underway right now.
Apart from economic concerns, Generative AI brings forth some profound societal challenges. The filmmaker said that it has the power to propagate misinformation and political propaganda on an unprecedented scale. It has the potential to sway elections and undermine the basic tenets of democracy itself. AI also encourages biases against minorities and women, and in some cases, it also resorts to deception to meet its goals.
According to the author, Generative AI cannot be entrusted with mission-critical operations such as industrial processes. He went on to share three things that everyone should remember about AI.
First, Generative AI doesn’t care about you, nor does any AI. Human life means nothing to AI, it’s just a program.
Second, astoundingly, not even the experts know how it works. “You cannot look inside the model to see what’s going on. It’s literally a black box.”
Third, Generative AI has unexpected emerging capabilities. To elaborate on this, Barrat said when ChatGPT, the most popular LLM right now, was created the makers did not know that it could translate or write programs. “These properties just came out. However, it made the makers proud and excited. But, it also made everyone else wonder, what else will emerge from this Pandora’s box?”
While discussing the risks associated with generative AI, Barrat shared about his book published in 2013, ‘Our Final Invention: Artificial Intelligence and the End of the Human Era’. The author shared that although he did not think that the developments would occur this fast, he anticipated the exponential growth of AI and how it will advance at a rapid pace until we reach something known as ‘intelligence explosion’.
Barrat shared that in his book, he wrote that humans cannot outpace super-intelligent machines. He believes that these machines could potentially be millions of times smarter than humans, making them uncontrollable and incomprehensible similar to ChatGPT. “It’s like a cockroach trying to grasp human desires; in this scenario, humans are the cockroaches, facing bleak odds against AI,” he said.
As machines are rapidly becoming more intelligent than humans, we are unprepared to control them, believes Barrat. He warned that unlike the technologies of the past, these can outsmart humans. So far, humans have steered the future because of their intelligence. Barrat contends that if superior intelligence emerges, it will dictate the future of humanity. The real contest according to him is between AI and humanity which will impact all future generations.
Talking about solving the crisis, Barett also touched upon how CEOs of major AI companies are currently advocating for AI regulation. He said while Google’s CEO desires a balanced oversight, other CEOs emphasised alignment with human values. However, the author questioned the sincerity of these calls, he stated that the CEOs may be evading blame or influencing the regulation. Regardless, he added that entities such as the EU with their aim for human-centric AI regulations, are offering some hope.
“Announcing that your product could destroy mankind, excites investors. They think that you’re about to release a stronger product,” said Barrat while commenting on the Generative AI race which he terms as a suicide race.
While talking about the potential solutions, Barrat listed three things that we can do as citizens of the world.
First, acquire an understanding of the main AI tools. See how they’re being developed. See how they’re being used. Encourage your political officials to take strong positions on AI safely.
Second, tell politicians that we need an international AI government or something like the International Atomic Energy Agency. We desperately need that, we need it now.
Third, If your company uses AI, or if your school uses AI, make sure it’s using it safely.
James Barrat has dedicated his career to introducing audiences to complex ideas in an engaging and thought-provoking way. Time Magazine once called him one of the “very smart people” who think artificial intelligence could bring the apocalypse. His groundbreaking book, ‘Our Final Invention: Artificial Intelligence and the End of the Human Era,’ has sparked debates worldwide.
