Big business has shown small firms what to do – and what not to do – with AI | Gene Marks
A US small-business columnist argues that large corporations have effectively acted as unwitting guinea pigs for artificial intelligence, allowing smaller firms to learn from their successes and costly mistakes without bearing the same financial risk. The piece contends that big business has shown small and mid-sized businesses (SMBs) where AI genuinely delivers value, such as software development and customer service, while also revealing pitfalls around cost, trust and public messaging that smaller firms are now avoiding.
The author highlights specific lessons drawn from corporate experience: AI is proving useful for coding, customer service, cybersecurity and voice-based systems, prompting some SMBs to hire developers laid off by big tech. Conversely, small firms have noted the huge computing costs some corporations incurred, the backlash CEOs faced after publicising AI-driven job cuts, and unreliable "agentic" AI tools, leading SMBs to favour cautious, productivity-focused AI adoption over sweeping automation or headcount reduction, while remaining sceptical of bold predictions about AI's transformative pace.
- Small firms are learning from big companies' AI wins and failures
- AI works well for coding, customer service, security and voice tasks
- SMBs avoid publicised layoffs, high compute costs and unreliable AI agents
Both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
Large companies’ experiments have shown that AI can deliver practical gains in areas such as coding, customer service and cybersecurity, while giving smaller firms evidence about which tools are useful. By adopting proven applications cautiously, small businesses may improve productivity without taking on the costs and risks of broad, untested automation.
The case against
Corporate examples may not provide a reliable guide for smaller firms, whose budgets, workloads and customer relationships differ, and AI tools can remain costly or unreliable even when they appear successful elsewhere. A cautious approach could also mean missing productivity gains or falling behind competitors as the technology improves.