Clues to the future of AI? Deutsche Bank study lists THESE five topics to watch

Deutsche Bank, in its research report on artificial intelligence titled ‘The latest in AI: Five clues for the future of AI’, Alabama In recent headlines, you have listed the topics to be considered in this space.

Noting that AI is “more fashionable than ever”, with AI and ChatGPT While technologies continue to dominate Google search trends, the report also recalled older trends that experienced a similar surge in popularity before fading away: virtual reality (VR), the metaverse and cryptocurrencies.

However, he added that searches for Al in September 2024 “are as high as they’ve ever been before, and 20 times higher than they’ve ever been before.” crypto“He added that five songs “give strong clues” about Al’s future for the rest of the year.

Five issues to consider

1. Al is the elephant in the room
2. Data centers are Ground Zero for Artificial Intelligence
3. Al remains a magnet for funding
4. Political appetite is growing for Al Regulation
5. Real-world uses have yet to emerge on a large scale.

First: Al is the elephant in the room.

The report noted that conversations about AI in the news are often tied to the performance of companies’ stocks. It said that while “The Magnificent Seven‘Tech stocks have soared and now account for 30 percent of the S&P 500’s $50 trillion total market capitalization, but it’s unclear whether big bets on AI are paying off.

The S&P’s roller coaster ride this year, with 14 percent growth in the first half of fiscal 25 followed by a flat third quarter, has been driven by big tech companies, including Nvidiawhich has bet heavily on its artificial intelligence chips. He noted that while investors are counting on valuations, they are also looking for long-term metrics to support the stock price.

“Yes, the Magnificent Seven stocks have outperformed the S&P 500 for 15 of the past 21 months… but their price-to-earnings ratioearnings “The earnings-to-earnings ratio is around 30%, well below the 90-times earnings-to-earnings ratio achieved by Microsoft and 160-times by Oracle during the dot-com bubble,” the report said.

It was also noted that returns are not uniform. Return on invested capital ranges from 15% for laggards, around 30% for the majority, and over 100% for Nvidia (the main gainer), raising the question of “overinvestment.”

Alphabet Chief Executive Sundar Pichai told analysts in July that the risk of under-investing is “dramatically greater” than that of over-investing, and Meta’s Mark Zuckerberg, on a Bloomberg podcast, acknowledged that hindsight can reveal overspending.

There are also concerns that Nvidia It is entirely dependent on “a handful of companies for sales” and the technology’s capability, value and reliability are not yet established.

Second: Data centers are ground zero for artificial intelligence

A fully functional AI needs supporting physical infrastructure, such as data centers (there are more than 10,000 of them around the world, each with thousands of servers running the nebulous “cloud”). The report noted that Black stone And the Canada Pension Plan Investment Board’s $2 billion investment plan in this space demonstrates the demand.

This, in turn, will also skyrocket energy demand as bigger and better AI programs will require larger-scale models and processors to improve and support the performance of various AI tools.

Examples? Dario Amodei, CEO of Amazon AnthropicZuckerberg, the creator of the Claude 3 chatbot, said that training today’s most advanced AI models will cost around $1 billion, and that training the next generation will cost ten times as much. In terms of GPUs, Meta’s latest large language model (LLM), Llama 3, was trained on two clusters of 24,000 Nvidia H100 chips, and Zuckerberg recently said Llama 4 will need ten times as much.

While data center power demand has remained stable at around 200 TWh over the past decade or so, demand is now increasing. According to the International Energy Agency, Amazon alone purchased more than 20 GW of renewable energy between 2010 and 2022, while Meta and Microsoft Each has purchased more than 10 GW. Earlier this year, Microsoft said its carbon emissions have risen by nearly 30 percent since 2020, largely due to data center construction, though it aims to be carbon neutral by 2030.

Three: Al remains a magnet for funding

Despite the doubts, AI remains attractive to investors and attracts “generous funding” in private markets. The biggest example of success of this is OpenAIwhich was founded with 10 employees, private capital and the goal of developing secure artificial intelligence systems that surpass human capabilities.

The money keeps flowing. In August, OpenAI said it is looking to raise billions of dollars at a valuation of over $100 billion; another example is Tokyo-based Sakana Al, founded only in July 2024, which reached unicorn status ($1 billion valuation) last week.

While global private investment in AI fell to $96 billion in 2023 overall, infusion into strategic sectors has increased: $18 billion was invested in AI. infrastructureresearch and governance last year, compared to $1 billion in 2022.

Four: Political appetite for AI regulation is growing

Calls for AI regulation are growing louder. California has a bill ready, called “SB 1047‘, and the European Union has its AI Law. It is worth noting that California, where Silicon Valley is located, is home to 32 of the 50 most important generative AI companies in the world.

The report notes that SB 1047, passed by state lawmakers on Aug. 29, reflects “a growing political appetite around the world to recognize the benefits while, in the words of the bill, avoiding the most serious risks and ensuring access.”

SB 1047, which contains provisions requiring companies operating in California to implement a number of safeguards before training advanced AI models, has been met with both criticism and approval. OpenAl, Google, and Meta have expressed opposition, while Anthropic, Musk And godfather-turned-conscience Geoffrey Hinton gave his blessing, the report noted.

He EU AI LawImposed on August 1 and most of whose provisions will come into force in two years, it is the world’s first horizontal legal framework regulating AI.

On the global stage, the report noted that in the absence of specific AI regulation, a cascade of antitrust and copyright cases is shaping the environment. There are also fears that AI will replace jobs, disrupt industries, affect security and the technology will be misused.

Five: Real benefits have yet to emerge on a large scale

Overall, despite early adopters, AI It has yet to demonstrate its real benefits on a large scale, the report said.

“The gap between high experimentation and low adoption is particularly striking in regulated sectors such as financial services and healthcare. In theory, they have the most to gain from accessing and analysing mountains of unstructured data, but in practice they also have more at stake than most,” the report notes.

He also acknowledged that Generative AI “It is certainly imperfect” and questions remain about whether productivity goals justify the means of investment. He also noted that technology historically takes years to implement and often depends on adjacent innovations.

The report, however, is optimistic: “Today, AI does not need to be perfect to be revolutionary,” it says.