The artificial intelligence (AI) gold rush is truly underway. After the release last November of ChatGPT—a game-changing content-generating platform—by research and development company OpenAI, several other tech giants, including Google and Alibaba have raced to release their own versions.
Investors from Shanghai to Silicon Valley are now pouring tens of billions of dollars into startups specialising in so-called generative AI in what some analysts think could become a new dot-com bubble.
The speed at which algorithms rather than humans have been utilised to create high-quality text, software code, music, video, and images has sparked concerns that millions of jobs globally could be replaced and the technology may even start controlling humans.
Also Read | ChatGPT: Disruption or fantasy?
But even Tesla boss Elon Musk, who has repeatedly warned of the dangers of AI, has announced plans to launch a rival to ChatGPT.
ChatGPT quickly adopted
Businesses and organisations have quickly discovered ways to easily integrate generative AI into functions like customer services, marketing, and software development. Analysts say the enthusiasm of early adopters will likely have a massive snowball effect.
“The next two to three years will define so much about generative AI,” David Foster, cofounder of Applied Data Science Partners, a London-based AI and data consultancy, told DW. “We will talk about it in the same way as the internet itself—how it changes everything that we do as a human species.”
Foster noted how generative AI is being integrated into tools companies already have, like Microsoft Office, so they do not need to make huge upfront investments to get a significant benefit from the technology.
ChatGPT and the others are still far from perfect, however. They mostly assist in the creative process with prompts from humans but are not yet worker substitutes. But last month, an even more intelligent upgrade, ChatGPT-4 was rushed out, and version 5 is rumoured for release by the end of the year.
Another advancement, AutoGPT, was launched at the end of last month, which can further automate tasks that ChatGPT needs human input for.
Billions pour into AI projects
Research last month by Deutsche Bank showed that total global corporate investment into AI has grown 150 per cent since 2019 to nearly $180 billion (€164 billion), and nearly 30-fold since 2013. The number of public AI projects rose to nearly 350,000 by end of last year, with more than 140,000 patents filed for AI technology alone in 2021.
Startups do not need to reinvent what has already been created. Instead, they can focus on adapting the current generative AI platforms for specialist uses, including cures for cancers, smart finance, and gaming.
“You have a new market emerging, a bit like when the [smartphone] app stores opened up. Small startups will make creative use of the technology, even though they didn’t create it themselves,” author and AI researcher Thomas Ramge told DW.
While the US has until now led the world in AI development, China has recently closed the gap along with India. China is now responsible for 18 per cent of all high-impact AI projects, compared to 14 per cent for the US, according to Deutsche Bank.
China steps up as AI competitor
The East-West race for economic dominance, however, is overshadowed by the threat of how an authoritarian government, like Beijing, could further use AI to control not only its population but the rest of the world. Some think this fear is overblown, however, as China’s leaders have their own anxieties over the power of algorithms.
“The Chinese government has been regulating AI because they see very clearly that it could cause them to lose control,” AI expert and MIT professor Max Tegmark told DW. “So they’re limiting the freedom of companies to just experiment wildly with poorly understood stuff.”
Tegmark is more concerned about the race by Western tech giants to push the technology toward the outer edges of acceptability and beyond. He noted that the US is hesitant to introduce AI regulations, due to lobbying by the tech sector. Repeated warnings about the need to avoid a so-called AI arms race have fallen on deaf ears.
“Sadly, that’s exactly what we have right now,” said Tegmark, “They [corporate leaders] understand the risks, they want to do the right thing, but they can’t stop. No company can pause alone because they’re just going to have their lunch eaten by the competition and get killed by their shareholders.”
Two years of work by the European Union on the Artificial Intelligence Act, which was due to be enacted this year, was upended by the launch of ChatGPT, which sent policymakers back to the drawing board.
Also Read | Privacy a casualty in race to develop generative AIs
Europe, meanwhile, is struggling to match the hunger of its US and Asian tech counterparts in the generative AI space due to investors being risk-averse. “Same old story. Europe is lagging behind,” Ramge said. “It did not foresee this trend and is once again claiming it will be able to catch up.”
Ramge highlighted two potential stars—a German plan to create a European AI infrastructure known as LEAM, and the Heidelberg-based startup Aleph Alpha, despite the latter raising just $31.1 million to date, versus OpenAI’s $11 billion.
“What Europe is not able to do is to transfer the knowledge out of the universities into rapidly growing startups — unicorns — that in the end are able to bring the new technology to the world,” he told DW.