AI is taking over 93% of US workplaces but only half are trained to use it: Who's falling behind, and why?
Artificial intelligence is no longer an experiment in the American workplace: it's the new normal. From retail to finance, healthcare to logistics, AI tools are transforming how companies operate.
But while the tools are rolling out rapidly, training for the people expected to use them isn't catching up. That growing disconnect is becoming one of the most overlooked risks in the AI revolution.
According to the
Forbes Research 2025 CxO Growth Survey
, 93% of US companies plan to increase their AI investments over the next two years. More than half (56%) are boosting their AI budgets by over 16%, making it one of the top two technology investment areas, alongside cybersecurity.
Yet only 49% of CHROs say their organisations are currently prioritising training in AI and data analysis. Even among companies with the most aggressive AI spending plans, that number only reaches 57%. And in the technology sector, where one might expect AI preparedness to be strongest, only 38% of HR leaders say AI training is a top priority.
The tools are ready, but are the people?
AI tools are already reshaping business operations. Sixty-nine percent of executives say AI agents are transforming automation, allowing teams to shift focus to more strategic work.
But strategic outcomes depend on human capability: and that's where many companies are falling short.
Without proper training, employees may lack the confidence or knowledge to use AI tools effectively. That leads to underutilisation, inefficient processes, and in some cases, failed implementations. The promise of AI often goes unrealised not because the technology is flawed, but because the people using it aren't equipped to do so.
Why companies are leaving training behind
Speed is a big factor. In the rush to implement AI tools and stay ahead of competitors, training is often deprioritised. There's also a common assumption that younger, digitally fluent employees will adapt on their own. But AI literacy is more than familiarity — it involves understanding how algorithms work, interpreting outputs, and knowing when to override or escalate decisions made by machines.
In many companies, training is ad hoc or reactive.
Employees are handed new tools and left to 'figure it out.' That might work for simple software updates. But with AI reshaping job roles, workflows, and decision-making, a more structured approach is needed.
The human cost of falling behind
When training lags behind tech, the consequences can ripple across the organisation. Employees may feel overwhelmed or anxious, unsure how to integrate new tools into their existing workflows. Teams may resist adoption entirely, especially if they fear job displacement.
And without the skills to question or verify AI outputs, mistakes can slip through — potentially damaging operations or customer trust.
The more AI becomes embedded in core business functions, the more critical it is that human users are brought along for the ride. Otherwise, the very systems designed to increase productivity could end up stalling it.
Closing the gap before it widens
The AI revolution in the US is moving fast, but it's not too late to bridge the readiness gap. Companies that pause to train, upskill, and engage their workforce may ultimately move faster and more successfully than those who rush to deploy without preparation.
As AI continues to expand into 93% of US workplaces, the question isn't whether people can keep up. It's whether companies will give them the tools, time, and training to do so.
TOI Education is on WhatsApp now. Follow us
here
.
Ready to navigate global policies? Secure your overseas future. Get expert guidance now!
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
&w=3840&q=100)
First Post
15 minutes ago
- First Post
Nvidia, AMD to give US 15% cut from China chip sales revenue
Nvidia and AMD have agreed to give the US government 15 per cent of their China chip sales revenue in exchange for export licenses, marking the first such deal by American companies, says report. read more Nvidia and AMD have agreed to give the US government 15 per cent of their revenue from chip sales in China in exchange for export licenses, the Financial Times reported, citing a US official. The deal covers Nvidia's H20 chips and AMD's MI308 chips. It marks the first time US companies have paid a portion of their revenues to secure export permissions. Licenses for the Chinese market were granted last week under the arrangement, though the Trump administration has not yet decided how to use the funds. STORY CONTINUES BELOW THIS AD Export control experts say the move is unprecedented. Trump has previously urged companies and countries to invest in the US to 'buy down' tariff rates. An Nvidia spokesperson told Reuters the company follows US rules for global markets. 'While we haven't shipped H20 to China for months, we hope export control rules will let America compete in China and worldwide,' the statement read. The US commerce department began issuing licenses for Nvidia's H20 exports to China last week, clearing a major obstacle to the company's access to a key market. The US had reversed its April ban on H20 sales last month, after Nvidia had redesigned the chip to meet Biden-era AI export restrictions.
Time of India
15 minutes ago
- Time of India
Goodbye, $165,000 tech jobs, student coders seek work at Chipotle
Academy Empower your mind, elevate your skills Since the early 2010s, a parade of billionaires, tech executives and even US presidents has urged young people to learn coding, arguing that the tech skills would help bolster students' job prospects as well as the economy. Tech companies promised computer science graduates high salaries and all manner of perks."Typically their starting salary is more than $100,000," plus $15,000 hiring bonuses and stock grants worth $50,000, Brad Smith, a top Microsoft executive, said in 2012 as he kicked off a company campaign to get more high schools to teach financial incentives, plus the chance to work on popular apps, quickly fed a boom in computer science education, the study of computer programming and processes like algorithms. Last year, the number of undergraduates majoring in the field topped 170,000 in the United States -- more than double the number in 2014, according to the Computing Research Association, a now, the spread of AI programming tools , which can quickly generate thousands of lines of computer code -- combined with layoffs at companies like Amazon, Intel, Meta and Microsoft -- is dimming prospects in a field that tech leaders promoted for years. The turnabout is derailing the employment dreams of many new computing grads and sending them scrambling for other college graduates ages 22 to 27, computer science and computer engineering majors are facing some of the highest unemployment rates, 6.1% and 7.5% respectively, according to a report from the Federal Reserve Bank of New York. That is more than double the unemployment rate among recent biology and art history graduates, which is just 3%.In response to questions from The New York Times, more than 150 college students and recent graduates -- from state schools as well as private universities -- shared their experiences. Some said they had applied to hundreds, and in several cases thousands, of tech jobs at companies, nonprofits and government computing education boosters are now pivoting to AI. President Donald Trump, who in 2017 directed federal funding toward computer science in schools, recently unveiled a national AI action plan that includes channeling more students into AI a major computing education sponsor, recently said it would provide $4 billion in technology and funding for AI training for students and workers. Last month, Smith said the company was also assessing how AI was changing computer science education.
The Hindu
15 minutes ago
- The Hindu
How artificial intelligence is tackling mathematical problem-solving
The International Mathematical Olympiad (IMO) is arguably the leading mathematical problem-solving competition. Every year, high school students from around the world attempt six problems over the span of three hours. Students whose scores cross a threshold, roughly corresponding to solving five of the six problems, obtain Gold medals, with Silver and Bronze medals for those crossing other thresholds. The problems do not require advanced mathematical knowledge, but instead test for mathematical creativity. They are always new, and it is ensured that no similar problems are online or in the literature. The AI gold medallist IMO 2025 had some unusual participants. Even before the Olympiad closed, OpenAI, the maker of ChatGPT, announced that an experimental reasoning model of theirs had answered the Olympiad at the Gold medal level, following the same time limits as the human participants. Remarkably, this was not a model specifically trained or designed for the IMO, but a general-purpose reasoning model with reasoning powers good enough for an IMO Gold. The OpenAI announcement raised some issues. Many felt that announcing an AI result while the IMO had not concluded overshadowed the achievements of the human participants. Also, the Gold medal score was graded and given by former IMO medalists hired by OpenAI, and some disputed whether the grading was correct. However, a couple of days later, another announcement came. Google-DeepMind attempted the IMO officially, with an advanced version of Gemini Deep Think. Three days after the Olympiad, with the permission of the IMO organisers, they announced that they had obtained a score at the level of a Gold medal. The IMO president Prof. Gregor Dolinar stated, 'We can confirm that Google DeepMind has reached the much-desired milestone, earning 35 out of a possible 42 points — a gold medal score. Their solutions were astonishing in many respects. IMO graders found them to be clear, precise and most of them easy to follow.' Stages of development Even as it became a popular sensation, ChatGPT was infamous both for hallucinations (making up facts) and for simple arithmetic mistakes. Both these would make solving even modest mathematical problems mostly impossible. The first advance that greatly reduced these errors, which came a few months after the launch of ChatGPT, was the use of so-called agents. Specifically, models were now able to use web searches to gather accurate information, and Python interpreters to run programs to perform calculations and check reasoning using numerical experiments. These made the models dramatically more accurate, and good enough to solve moderately hard mathematical problems. However, as a single error in a mathematical solution makes the solution invalid, these were not yet accurate enough to reach IMO (or research) level. Greater accuracy can be obtained by pairing language models with formal proof systems such as the Lean prover — a computer software that can understand and check proofs. Indeed, for IMO 2024 such a system from Google-DeepMind called AlphaProof obtained a silver medal score (but it ran for two days). Finally, a breakthrough came with the so-called reasoning models, such as o3 from OpenAI and Google-DeepMind's Gemini-2.5-pro. These models are perhaps better described as internal monologue models. Before answering a complex question, they generate a monologue considering approaches, carrying them out, revisiting their proposed solutions, sometimes dithering and starting all over again, before finally giving a solution with which they are satisfied. It were such models, with some additional advances, that got Olympiad Gold medal scores. Analogical reasoning and combining ingredients from different sources gives language models some originality, but probably not enough for hard and novel problems. However, verification either through the internal consistency of reasoning models or, better still, checking by the Lean prover, allows training by trying a large number of things and seeing what works, in the same way that AI systems became chess champions starting with just the rules. Such reinforcement learning has allowed recent models to go beyond training data by creating their own synthetic data. The implications Olympiad problems, for both humans and AIs, are not ends in themselves but tests of mathematical problem-solving ability. There are other aspects of research besides problem-solving. Growing anecdotal experiences suggest that AI systems have excellent capabilities in many of these too, such as suggesting approaches and related problems. However, the crucial difference between problem-solving and research/development is scale. Research involves working for months or years without errors creeping in, and without wandering off in fruitless directions. As mentioned earlier, coupling models with the Lean prover can prevent errors. Indications are that it is only a matter of time before this is successful. In the meantime, these models can act as powerful collaborators with human researchers, greatly accelerating research and development in all areas involving mathematics. The era of the super-scientist is here. Siddhartha Gadgil is a professor in the Department of Mathematics, IISc
