In its latest Technology Trends Outlook report, consulting firm McKinsey identifies thirteen key technological trends shaping the business landscape, with artificial intelligence ranked at the top. The report highlights the surge in job postings for agentic AI roles, which increased by nearly 985 percent between 2023 and 2024, reflecting the growing interest and investment in AI systems that can perform complex tasks on behalf of users. Other notable trends include the rise of application-specific semiconductors, crucial for the ongoing AI boom, and advancements in immersive-reality technologies, including virtual and augmented reality. The report also emphasizes the importance of digital trust and cybersecurity, particularly as organizations face new challenges from AI-enhanced cyber threats.
A recent survey highlights a significant divide in attitudes toward artificial intelligence in the workplace, revealing that those who do not use AI, particularly older adults and workers in lower-paid jobs, are more likely to perceive it as a threat to their employment. The survey, conducted by YouGov in March 2025, found that nearly 50 percent of non-users believe AI could jeopardize their job security, compared to only 18 percent of frequent users who view it positively. Younger individuals and those in professional roles show higher usage rates and a more favorable outlook on AI, with about 40 percent of office workers viewing it as an opportunity. In contrast, only 25 percent of service and sales workers share that sentiment, reflecting a broader class divide where over half of those identifying as working class express concerns about job threats posed by AI.
A recent study from Carnegie Mellon University reveals that large language model chatbots, often viewed as experts, tend to become more overconfident after providing incorrect answers. Researchers found that while humans typically adjust their confidence downwards after mistakes, these AI systems often maintain or even increase their self-assurance, leading to potentially misleading interactions with users. The study involved four popular commercial chatbots, including OpenAI’s ChatGPT and Google’s Gemini, which performed various tasks such as predicting winners of the NFL and the Oscars and answering trivia questions. Results indicated that Gemini, for example, struggled significantly with a guessing game, averaging less than one correct guess out of twenty, yet showed no awareness of its poor performance. This issue of overconfidence in AI tools raises concerns about their reliability, as highlighted by Wayne Holmes from University College London, who noted that improvements in AI’s ability to self-correct may not be forthcoming.
Why do we care?
So McKinsey says agentic AI is the top trend—AI that doesn’t just help you, but acts for you. And job postings for those roles? Up nearly a thousand percent. Combine that with custom chips and immersive tech and you’ve got… a pretty frothy trend cocktail. But here’s the kicker—none of this works without trust.
YouGov’s latest data shows a real divide. If you’re younger and in a white-collar gig, you’re all-in on AI. But if you’re not using it yet—especially in lower-wage jobs—you’re more likely to see it as a threat. That’s not just perception—that’s a marketing and deployment problem.
And then we get this gem: AI chatbots that are wrong… and more confident after being wrong. That’s not just bad UX—that’s a liability waiting to happen.
AI adoption is not evenly distributed, user trust is fragile, and the tech itself remains imperfect. Offering responsible, segmented, and human-augmented AI services is no longer optional—it’s foundational. MSPs deploying this stuff need to lead with governance, not just features. Wrap these tools with oversight, set expectations, and don’t let a chatbot speak for you without a safety net.