Executives on whether ‘learn to code’ is still good advice in AI era
Summary
Business Insider asked eight tech executives whether the long-standing advice to “learn to code” still holds up as AI automates more of software development. Responses vary: some leaders — including Google’s head of research and Cloudflare’s CEO — say basic coding knowledge remains valuable, especially for understanding systems and reviewing AI output. Others, like Autodesk’s CEO and Salesforce’s AI EVP, argue that systems-level thinking, creativity and “agency” are becoming more important than writing code. Okta’s CEO went further, calling the idea that everyone must learn to code “laughable.”
Author’s take (punchy): Quick read, useful consensus: basics matter, but the skill mix is shifting — if you’re not wired for detailed coding, focus on systems thinking and how to orchestrate AI.
Source
Source: https://www.businessinsider.com/executives-weigh-in-on-learn-to-code-advice-2025-9
Key Points
- AI already generates a substantial share of code at major firms (20–30% reported by Microsoft and Google execs).
- Many executives say basic coding knowledge is still useful for understanding systems and evaluating AI outputs (Cloudflare, Google, Cisco, Bluesky, Dropbox).
- Some leaders prioritise systems-level thinking — understanding how people, processes and tech connect — over pure coding (Autodesk).
- Having “agency” — the drive to define problems and orchestrate solutions — is argued to be more critical than coding for many roles (Salesforce).
- Security and quality concerns mean human review remains essential; AI-generated code is not a full substitute (Cloudflare emphasis).
- Not everyone should be pushed into coding; aptitude and role fit matter (Okta CEO’s view).
- Product and non-engineering roles benefit from knowing the tech stack and some SQL or basic coding to spot opportunities (Dropbox).
- Executives broadly recommend a hybrid approach: learn the basics, then specialise in higher-level thinking or orchestration if coding isn’t your strength (consensus view).
Context and Relevance
This debate matters for career decisions, hiring and education. As AI tools automate routine development tasks, employers increasingly prize people who can design systems, evaluate AI outputs and connect technology to business problems. For learners and professionals, the takeaway is to balance technical foundations with skills in problem framing, cross-disciplinary thinking and leadership of AI-driven workflows.
Why should I read this?
Short version: yes — skim this if you’re wondering whether to bother with coding. We’ve done the reading for you: the execs mostly say learn the basics (so you can tell good work from rubbish), but don’t freak out — if coding doesn’t suit you, double-down on systems thinking, product sense and the ability to get AI to do useful work. In plain terms: don’t panic, adapt your skillset.
Source
Source: https://www.businessinsider.com/executives-weigh-in-on-learn-to-code-advice-2025-9