What: the Economist reviews why AI is not poised to boost productivity in the immediately coming years.

Why is it important: Where does your company belong in its relationship to AI? window dressing, low-killed workers reskilling, or valuable employees enhancement?

The article discusses how businesses across sectors are starting to adopt AI tools like ChatGPT, but adoption remains low overall. As of February 2024, only about 5% of US firms were using AI, with the highest adoption in tech and information companies (17%) and lower levels in sectors like manufacturing (3%). Big tech firms like Microsoft, Google, and Amazon are seeing some revenue growth from AI offerings, but it still accounts for a small fraction of their cloud and services sales. For the tech stock boom to continue, these companies need businesses to start utilizing AI on a large scale. Most companies are still just piloting AI or only have limited use cases compared to full integration into operations. Common early experiments fit into three categories: rebranding digital efforts as "AI transformation", using AI for simple customer service chatbots and administrative tasks like data entry, and deploying tools aimed at high-value workers for functions like writing code and improving search. Challenges holding back further adoption include lack of internal skills, data too fragmented, legal and ethical concerns around data privacy and AI mistakes, and the current limitations of AI tools in terms of accuracy and capabilities. As examples, AI is helping some teams achieve 10-20% productivity gains in areas like software development. But issues like bugs in AI-written code can create other inefficiencies. Adoption is steadily increasing as companies build expertise and AI technology improves, but it will take time for measurable economy-wide productivity impacts.

How businesses are actually using generative AI