ChatGPT is two years old. Here’s a look at the state of the generative AI union.

OpenAI’s ChatGPT celebrates its second year of general availability on November 30, which means it’s time to take stock of the progress of generative AI in 2024. And what a whirlwind it was.

Two years ago, few people knew GenAI was possible, but today about 40% of Americans use the technology. To put this adoption trajectory in perspective, according to the National Bureau of Economic Research, this growth doubles the 20% of Americans who used the Internet within two years of its launch.

You read that right: The damn internet.

The pace of GenAI innovation has been unprecedented. In 2024 alone, OpenAI broke new ground in LLM reasoning while Meta introduced the first open frontier class model. Google has now made a breakthrough in GenAI-powered podcasting.

And Anthropic launched tools to help users create and modify content in a separate window, as well as the ability for computers to use computers. (You have to see it to understand).

AI agents shine brightly

The excitement around AI agents is palpable as companies look to increase not only employee productivity but also operational efficiency. At a higher level, AI agents are pieces of software code that perform tasks to achieve a given goal. Most AI agents can “think,” reason, plan, and learn from feedback.

However, agents can also take many forms. First, AI agents can include digital assistants that help consumers. Think about software bots that can book trips and handle other transactions, etc. Then there are enterprise agents that can work individually or as part of teams (multi-agent architectures) to automate entire workflows or entire business processes. Ultimately, these agents will be able to heal themselves, identify errors, and course correct.

And while it’s premature to claim that AI agents will automate an entire company, companies are certainly interested in their potential. 82 percent of executives surveyed by Capgemini said they expect to use agents to automate email generation, software code and data analysis.

Small language models can achieve big things

Some people are concerned about the money hyperscalers are pumping into GenAI’s infrastructure, software and talent. However, you understand it Motivation is critical; These companies are investing in super-intelligent systems – a huge leap beyond the everyday content creation applications that most companies pursue.

The reality is that companies don’t need to spend millions of dollars building or licensing large language models (LLMs). Rather, small language models (SLMs) running in hybrid IT environments provide more than enough AI firepower to cover most targeted business use cases.

“You’ll see a number of use cases where a small, less accurate model is much better than what you had, and probably good enough,” Mindy Cancila, vice president of corporate strategy at Dell Technologies, told a recent webinar.

Additionally, the smaller footprint of SLMs means they can run on everything from servers to laptops to smartphones, and be powered by data stored anywhere from corporate data centers to public cloud services to the edge, where breakthroughs Model compression and performance will enable high-quality inferences with low latency.

Progress will bring with it loss of productivity

Numerous studies suggest that GenAI has increased productivity across companies. In reality, actual results are difficult to quantify, according to scientist Ethan Mollick, an expert in adopting GenAI in organizations, who found that business leaders report low AI adoption and few productivity gains outside of niche use cases.

Mollick also argues that companies need to conduct research and development to understand the productivity of AI usage and other metrics of progress. And these R&D analyzes have yet to be codified – even by the consulting firms that are paid to do the work.

“No one has specific information about how to best use AI in your business or a guide for integrating it into your business,” Mollick said.

Nevertheless, consulting firms continue to find positive metrics from GenAI investments and adoption.

For example, Ernst & Young LLP found that executives whose current AI investment budgets accounted for 5% or more of their total budget achieved higher positive returns in several key areas than those who spent less than 5%.

According to EY, those who allocated 5% or more of their budget outperformed their lower-spending peers by 76% to 62% in employee productivity, 71% to 55% in product innovation, and 73% to 47% in creating competitive advantage.

This suggests that companies can ill afford not to increase investments in GenAI. Your competitors certainly will.

Of course, aligning business strategy with IT investments to achieve desired business outcomes is never easy. But you don’t have to do it alone; Trusted advisors like Dell are here to help.

Find out more about the Dell AI Factory.