This set of Friday Big Ideas has a bit of a theme to consider.
In Business Insider, Salesforce Chief Executive Officer Marc Benioff has stated that up to 50% of the company’s work is currently being performed by artificial intelligence. This raises significant questions about the future of Salesforce’s workforce, which numbered over 76,000 employees as of January 2025. … and In a recent discussion, Senator Bernie Sanders suggested that if artificial intelligence is as transformative as claimed, then workers should benefit from a reduced 30-hour workweek. Sanders argues that technology should enhance the lives of all workers, not just corporate executives, emphasizing the need for more time for personal pursuits and family.
But this may not be all great, as the New York Times highlights how Artificial intelligence is increasingly undermining democratic processes around the world, with evidence from over 50 countries indicating its disruptive influence in elections. A report by the International Panel on the Information Environment found that in 2024, artificial intelligence played a significant role in more than 80 percent of elections surveyed, with 215 documented instances of manipulation.
And it may not be great in education either. A recent study from the Massachusetts Institute of Technology’s Media Lab indicates that students who relied on ChatGPT for writing tasks exhibited lower brain engagement and creative output compared to their peers who wrote without assistance. In the study, 54 participants aged 18 to 39 were divided into three groups: one used ChatGPT, another used Google Search, and the last wrote without any tools. The ChatGPT group produced essays that lacked originality and depth, with researchers noting a decline in engagement over time. In a related note, a study from Pew Research indicates that around 25% of U.S. teenagers have utilized AI tools for schoolwork, reflecting a growing trend in educational technology. The paper is still awaiting peer review.
Second-order effects include unexpected impacts, and here’s one I didn’t anticipate. Fanfiction writers are coming together to defend their work from being used without permission. Recently, over 12 million fanfiction stories were scraped from the online repository Archive of Our Own and uploaded to a public dataset, causing outrage among the fan community. Fanfiction writers, who usually operate within a gift economy and do not make a profit from their work, argue that their creations are being exploited to train generative AI models. The Organization for Transformative Works, which manages Archive of Our Own, has taken steps to protect user content, but problems remain, as some scrapers continue to upload fanfiction to less regulated platforms.
But I won’t make it all a bummer.
From Information Week. Shadow IT, often seen as a security risk, might actually act as a driver for innovation within organizations. As unauthorized technology use grows, especially with artificial intelligence applications, IT departments are encouraged to shift their focus from stopping it to managing it. A recent report by Skyhigh Security showed that companies are using over 320 unapproved AI applications, with traffic to these tools increasing by 200% last year. Additionally, a survey from Harness indicated that only 48% of developers use IT-approved AI tools, revealing a large gap between business needs and technology solutions. Experts recommend that organizations should view shadow IT as an opportunity to find and expand effective tools, rather than imposing strict rules that could worsen covert usage. By creating an environment where employees can use innovative tech solutions, businesses can improve resilience and gain a competitive edge in the fast-changing digital world.
Why do we care?
Benioff’s comment is the tell: We’re still in the AI-hype phase, where big claims are designed to stimulate demand, not reflect reality. If 50% of your work is AI-driven and nothing changes structurally in your company—either you’re overstating impact or hoarding gains.
Sanders’ position, while politically loaded, is more practical: If AI raises productivity, it must be translated into time, compensation, or wellbeing for workers too. Otherwise, backlash is inevitable.
Because unlike Benioff’s claims, The New York Times reports that AI disrupted more than 80% of elections globally last year. That’s not hyperbole—it’s based on 215 documented manipulation incidents across 50 countries. AI can enhance operations—but unregulated use actively threatens stability.
While the MIT paper is pending peer review, it highlights a critical question for educators and employers alike: Are we training a generation to think less deeply?
The fan fiction example isn’t a fringe issue. It’s a bellwether for future content disputes across industries. For IT service firms managing compliance or advising on ethical data use, client policies around AI data sourcing are about to become a lot more relevant.
For IT service providers, this underscores two imperatives:
- Build trust by helping clients manage the messy reality of AI—ethics, compliance, creativity loss, and shadow tools.
- Push back on inflated AI vendor claims. Your credibility comes from honest assessment, not hype adoption.