The widespread use of AI tools is not boosting profits
MIT's report revealed that only about 5% of integrated AI pilots offer millions of value, but according to a report from the Daily Adda, there was no impact on revenue or revenue for the majority.
Many companies quickly began testing programs such as ChatGpt, Copilot and other large language models, with research showing that over 80% of major companies researching or piloting these tools, and reporting that they deployed to some extent 40%. However, most of this use is limited to helping individuals work faster than improving the profits of the entire enterprise, as reported by the daily ADDA.
Also Read: Does chatgpt help you get out of debt? What experts and users say about AI Chatbots financial advice
AI models can't think or learn like humans yet
One important issue is that, reports show that generative AI tools often don't fit into the real work process well. MIT research explains that AI tools, according to Daily Adda, are “fragile workflows, lack of context learning, and poor consistency with daily operations.”
Unlike humans, most generative AI models are unable to retain past feedback or build inference skills over time, and reports show that they struggle to adapt to new contexts or apply lessons across different tasks, and MIT's research shows that “most genai systems do not retain feedback, adapt to context, or improve over time” have long-term integrations so that these long-term integrations are not reported. The report says that AI has raised high expectations in the boardroom, but so far, investments have not led to better profits or meaningful cost savings. While some companies use AI to help with customer service, marketing, or writing, these tools save workers time, primarily without adding directly to their business revenue, according to daily ADDA reports.
AI could reduce outsourcing rather than work
Additionally, MIT's research downplays the fear of widespread unemployment due to generation AI in the near future, suggesting that the impact of AI is about reducing external costs for companies, such as outsourcing, rather than reducing the large number of jobs.
The study emphasized that “until AI systems achieve context adaptation and autonomous operation, as cited by Daily Adda, organizational impacts manifest themselves through external cost optimization rather than internal restructuring.”
Experts note that many failures have arisen from misunderstanding what AI can and cannot do to allow AI to generate text and code quickly, but reports show that they cannot actually learn how to do human beings. Meanwhile, employees can adjust based on new instructions, past mistakes, and changing needs, but the generated AI model cannot carry memory throughout the task unless retrying, as reported by daily ADDA.
Also Read: Tech stocks fall as investors escape – Is the AI bubble finally about to burst?
False expectations drive business AI failure
Despite ongoing interest from investors and executives, the short-term outlook shows that more progress will be slower than expected, and the study suggests that companies should view generative AI as a limited tool, rather than a guaranteed driver for growth.
Where is AI the most effective?
The study recommends focusing on narrow use cases where AI can provide immediate, measurable savings and productivity gains, such as customer support scripts, coding AIDS, or drafting documents.
FAQ
Have most companies seen business returns from AI?
95% of companies report no measurable profits from AI adoption or revenue, according to daily ADDA reports.
Does AI take on people's jobs?
It's not immediately. The report states that AI is likely to reduce external costs, such as outsourcing.
