Latest news with #DanielleLi
Hindustan Times
a day ago
- Business
- Hindustan Times
How AI agents could reshape the economy
Imagine a world where your digital assistant not only schedules meetings and books trips but also negotiates with dozens of businesses on your behalf, finding the best prices, customising services, and handling payments in seconds. This isn't science fiction. It's the emerging reality of what experts call the agentic economy, powered by generative AI. AI(REUTERS) Already, research from leading economists shows that these technologies are transforming productivity. A recent large-scale field experiment published in the Journal of Econometrics by Erik Brynjolfsson, Danielle Li, and Lindsey Raymond found that generative AI significantly boosts the output of knowledge workers, especially helping less-experienced employees reach the performance levels of their more seasoned peers. This real-world evidence offers a glimpse into how AI might not only make individuals more efficient but also help narrow skill gaps in the workforce. The concept of the agentic economy goes beyond individual productivity. As explored in a study by researchers at Microsoft Research, AI is poised to reduce the friction of communication between consumers and businesses. Instead of painstakingly filling out forms or explaining your situation repeatedly to different service providers, your assistant agent could securely transmit your preferences and history to business-side service agents, instantly tailoring offers to your needs. This could unlock entirely new options that were previously buried under cumbersome processes. For instance, switching accountants or trying a new insurer might become as easy as a short prompt to your AI assistant, who negotiates terms with multiple providers in real time. Such advances could dramatically alter who holds power in the digital economy. Today, giants like Amazon, Google, and Meta act as middlemen—platforms that set rules, curate experiences, and take commissions by bringing together millions of buyers and sellers. But if AI agents on both sides can communicate directly, the need for these costly intermediaries may shrink. Consumers' assistant agents could interact directly with businesses' service agents, comparing prices, customizing bundles, and even resolving disputes. This would foster a more decentralized, competitive market landscape. Still, experts caution that platforms often provide more than just matchmaking. They add value through validation, fraud prevention, and standardised experiences. As a result, we may see platforms evolve rather than disappear, competing fiercely in a market with much lower switching costs. Perhaps the most critical question is whether this agent-driven economy will flourish inside closed ecosystems, agentic walled gardens, or thrive in an open web of agents. Large tech companies are already taking steps to build their own controlled agent marketplaces. Meta, for example, recently launched business service agents on Facebook and Instagram that only work within their platforms. This model can ensure quality and security, but risks consolidating power in the hands of a few players, potentially stifling innovation and fragmenting the user experience. Conversely, an open agent ecosystem would resemble today's world wide web, where any consumer's agent could connect with any business's agent. This would democratise access and spur competition—but would require global cooperation on technical standards, along with robust systems for trust and security. AI agents could also revolutionise advertising. Right now, businesses pay to capture our attention. In an agentic economy, attention may be less scarce; instead, algorithms matching consumer assistants with service agents become the critical battleground. Paid prioritisation will likely remain in some form, but the true driver of success could be human feedback. As Brynjolfsson and his colleagues found, AI tools are most effective when they learn from high-quality user interactions. In the future, businesses may compete to attract early users whose feedback helps train smarter systems, shifting us from an attention economy to a preference economy. Imagine paying small amounts to access only what you need—whether it's a custom-tailored news article that skips what you've already read or a playlist dynamically mixed across streaming platforms. As assistant agents seamlessly handle transactions, micro-payments that once seemed impractical could become commonplace. This also sets the stage for extreme unbundling and rebundling of products. Assistant agents might pull from multiple content or service sources to build hyper-personalised offerings, negotiating micro-transactions behind the scenes. We stand at a pivotal moment, much like the dawn of the internet in the 1990s. Whether this next wave of AI delivers widespread opportunity or concentrates power even further depends on decisions made now by tech leaders, regulators, and consumers alike. The evidence is already here: From field experiments proving how AI boosts worker productivity to new frameworks enabling agents to communicate on our behalf. As we step into this agentic future, we must carefully choose the architecture of our digital economy because it will determine who benefits from this revolutionary technology. Generative AI is not just about personal productivity; it's about reshaping how markets work. Whether we end up in walled gardens controlled by tech giants or a vibrant, open web of competing agents will decide if this new economy truly serves us all. This article is authored by Narinder Kumar, assistant professor, RV University), Amit Kumar, research scientist, PGIMER, Chandigarh and Kiran Sood, professor, Chitkara University, Punjab.
Jordan Times
06-02-2025
- Business
- Jordan Times
Three reasons why AI's momentum could stall in 2025
LONDON – The rapid pace of technological advances over the past year, especially in artificial intelligence, has provided many reasons for optimism. But as we head into 2025, there are signs that AI's momentum may be waning. Since 2023, the dominant narrative has been that the AI revolution will drive productivity and economic growth, paving the way for extraordinary technological breakthroughs. PwC, for example, projects that AI will add nearly $16 trillion to global GDP by 2030, a 14 per cent increase. Meanwhile, a study by Erik Brynjolfsson, Danielle Li, and Lindsey R. Raymond estimates that generative AI could boost worker productivity by 14 per cent on average and by 34 per cent for new and low-skilled workers. Recent announcements by Google and OpenAI seem to support this narrative, offering a glimpse into a future that not long ago was confined to science fiction. Google's Willow quantum chip, for example, reportedly completed a benchmark computation, a task that would take today's fastest supercomputers ten septillion years (ten followed by 24 zeros), in under five minutes. Likewise, OpenAI's new o3 model represents a major technological breakthrough, bringing AI closer to the point where it can outperform humans in any cognitive task, a milestone known as 'artificial general intelligence.' But there are at least three reasons why the AI boom could lose steam in 2025. First, investors are increasingly questioning whether AI-related investments can deliver significant returns, as many companies are struggling to generate enough revenue to offset the skyrocketing costs of developing cutting-edge models. While training OpenAI's GPT-4 cost more than $100 million, training future models will likely cost more than $1 billion, raising concerns about the financial sustainability of these efforts. To be sure, investors are eager to capitalise on the AI boom, with venture capital firms investing a record $97 billion in US-based AI startups in 2024. But it appears that even industry leaders like OpenAI are burning through cash too quickly to generate meaningful returns, leading investors to worry that much of their capital has been misallocated or wasted. A back-of-the-envelope calculation suggests that a $100 billion investment in AI would require at least $50 billion in revenue to produce an acceptable return on capital, accounting for taxes, capital expenditures, and operating expenses. But the entire sector's annual revenues, according to my sources, total just $12 billion, with OpenAI accounting for roughly $4 billion. In the absence of a 'killer app' for which customers are willing to pay substantial sums, a significant portion of VC investments could end up worthless, triggering a decline in investment and spending. Second, the enormous amounts of energy required to operate and cool massive data centers could impede AI's rapid growth. By 2026, according to the International Energy Agency, AI data centers will consume 1,000 terawatt-hours of electricity annually, exceeding the United Kingdom's total electricity and gas consumption in 2023. The consultancy Gartner projects that by 2027, 40 per cent of existing data centers will be 'operationally constrained' by limited power availability. Third, large language models appear to be approaching their limits as companies grapple with mounting challenges like data scarcity and recurring errors. LLMs are primarily trained on data scraped from sources such as news articles, published reports, social media posts, and academic papers. But with a finite supply of high-quality information, finding new datasets or creating synthetic alternatives has become increasingly difficult and costly. Consequently, these models are prone to generating incorrect or fabricated answers ('hallucinations'), and AI companies may soon run out of the fresh data needed to refine them. Computing power is also approaching its physical limits. In 2021, IBM unveiled a two-nanometer chip, roughly the size of a fingernail, capable of fitting 50 billion transistors and improving performance by 45 per cent compared to its seven-nanometer predecessor. While undeniably impressive, this milestone also raises an important question: Has the industry reached the point of diminishing returns in its quest to make ever-smaller semiconductors? If these trends persist, the current valuations of publicly traded AI companies may not be sustainable. Notably, private investment is already showing signs of declining. According to the research firm Preqin, VC firms raised $85 billion in the first three quarters of 2024, a sharp drop from the $136 billion raised during the same period in 2023. The good news is that should today's AI giants start to falter, smaller competitors could seize the opportunity and challenge their dominance. From a market standpoint, such a scenario could foster increased competition and reduce concentration, preventing a repeat of the conditions that allowed the so-called 'Magnificent Seven', Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia, and Tesla, to dominate the US tech industry. Dambisa Moyo, an international economist, is the author of four New York Times bestselling books, including 'Edge of Chaos: Why Democracy Is Failing to Deliver Economic Growth – and How to Fix It' (Basic Books, 2018). Copyright: Project Syndicate, 2025.
