On April 4, Amazon Web Service, the cloud subsidiary of Amazon, announced that it has created a 10-week programme for generative artificial intelligence (AI) startups around the globe.
According to the company, generative AI startup founders within the cohort can expect to have access to AI models and tools as well as machine learning stack optimisation and custom go-to-market advice. There’s also a Demo Day in San Francisco at the end of the programme.
The development gives Amazon a different route to the frenetic AI party. Rather than bother with the hassles of building an energy-guzzling AI that may not be as good as OpenAI’s ChatGPT, why not own the companies already building something? That way Amazon can take advantage should any of the startups come up with a ChatGPT-like invention or something even better.
But companies like Nvidia are not leaving anything to chance. In March 2023, the United States-based chips and computer hardware manufacturer announced the release of its generative AI, marking the company’s largest-ever business model expansion.
While Microsoft and Google are the early comers to the party, companies like Adobe and Canva also threw their hats into the game in March.
Nevertheless, experts say Nvidia’s entrance is particularly exciting because generative AI requires intense computing power in data centres, which is a capability that Nvidia has already built up from its graphics processors and software designed for AI applications. Shortly after the company’s announcement, its stock rose 1 percent to close at $264.68. Year-to-date, Nvidia’s stock has risen to 81.1 percent.
The stakes in generative AI are rising. A report by Pitchbook finds that venture capitalists have increased investment in generative AI by 425 percent since 2020 to $2.1 billion. The surge in interest is also being aided by investors’ waning interest in Web3, thanks to Mark Zuckerberg’s Metaverse division losing $13.72 billion in investment. In addition, investors looking for the next hype following a lull in broader venture capital investment activity are also flocking to generative AI.
On March 7, Salesforce Ventures, the venture arm of the cloud software giant Salesforce, announced a $270 million generative AI investment fund. The fund has already been deployed to four generative AI startups: You.com, Anthropic, Cohere, and Health AI.
Startups like Stable Diffusion received $100 million of Series A funding on a $1 billion post-valuation led by Coatue Management, Lightspeed Venture Partners, and O’Shaughnessy Ventures in October 2022.
A new report by Global Data also acknowledges the growing investment in AI startups.
“The emergence of ChatGPT served as a pivotal moment for the technology industry, signalling the need for increased innovation and prompting many prominent leaders to expedite the development of their tools,” Pranjali Mujumdar, disruptive tech analyst at GlobalData, said. “Major tech companies such as Google, Microsoft, Meta, and Salesforce are investing heavily in generative AI research in hopes of achieving a breakthrough that would put them at the forefront of the race.”
‘Why generative AI? Why not generative AI?’
Behind the investments into generative AI is the race to control the future of content, one driven by smart machines and not humans.
Generative AI models leverage AI technologies to create entirely new content. Experts say this advancement from traditional machine analysis or AI exemplifies how AI programs have learned from large language models created by Google, Microsoft, and Meta. These large AI models are trained on vast quantities of data at scale and use large computing power resources to derive foundation models that are enabling the next phase of growth in artificial intelligence.
ChatGPT, which was developed by OpenAI in 2015, is the first commercialised generative AI that allows users to enter written prompts and receive new human-like text or images and videos.
“It is almost as if you are talking to a human. There are opportunities and challenges; one of which is the destruction of the conventional education system. It reduced the barrier to entry in most industries. It can even write code for you,” said Temitope Osunrinde, vice president, sales and marketing at Tizeti, an internet service provider.
ChatGPT is the most popular generative AI in the world, and even in Nigeria. Ahead of the presidential election in February, a United Kingdom-based fact-checking organisation, Full Fact, partnered with several African fact-checking agencies operating out of Nigeria. The company provided its AI suit, consisting of three tools that work in unison, to automate lengthy fact-checking processes.
“The tools are not intended to replace fact-checkers and the important work that they do,” said Kate Wilkinson, senior product manager at Full Fact. “Rather, they assist with time-consuming, manual monitoring and reviewing. This leaves fact-checkers more time to do the things they’re best at understanding what’s important in public debate, interrogating claims, reviewing data, speaking with experts, and sharing their findings.”
There have also been different experiments conducted by Nigerians with ChatGPT, with results posted on social media. Essentially, growing consumer interest in the capabilities of generative AI is behind the big-spending frenzy. ChatGPT debuted on November 30, 2022: it crossed 1 million users in less than one week, according to Sam Altman, CEO of OpenAI.
Read also: ChatGPT is fastest app to hit 100m users in history
What investors are betting on
As exciting as Microsoft’s ChatGPT, Google’s Bard and others may be, investors acknowledge they are still largely new technologies, and despite their potential, anything can go wrong.
ChatGPT, which seems to have everything going for it, still returns wrong answers. There is also uncertainty hanging over what approach regulation will take.
Nonetheless, experts identify three areas investments are going to: platforms such as ChatGPT or Bard; platform companies such as Microsoft or Google, or Nvidia; and app companies like Adobe. John Fortt, co-author of CNBC’s Closing Bell, said of the three, platform companies present the most opportunity for return on investment. This is because they can invest in AI applications and have the leverage to benefit when startups leverage their AP investments for their services.
For example, the new Nvidia AI Foundation services that span language, images, video, and 3D will be used by companies such as Getty Images, Morningstar, Quantiphi, and Shutterstock. Enterprises can also use the NVIDIA NeMo language service and the NVIDIA Picasso image, video, and 3D service to build proprietary, domain-specific, generative AI applications for intelligent chat and customer support, professional content creation, digital simulation, and more.
But the rising stakes have many countries worried. As AI gets more powerful and acquires human-like capabilities, millions of jobs are hanging in the balance and it doesn’t look good in the books of many governments. In April, Italy became the first country to officially ban ChatGPT over privacy concerns. Italy’s ban has attracted other members of the European Union like Germany, which is reportedly considering banning ChatGPT over privacy as well.
“AI bans are only treating the symptoms though and will stymie progress and science and cement monopolies elsewhere (or lead to abuse by the government instead). They should eliminate the profits on AI if they want to keep progress and society. Taxing direct and indirect profits from AI at 100 percent would level the playing field and fix the fundamental cause of most dangers with AI,” said Anders Edwards, a tech developer.
