Software companies adopted AI before anyone else, mostly from the bottom up. We pulled the developer surveys (Stack Overflow, GitHub, DORA) and the CIO-side data (Gartner, KPMG, Census, IBM) to show what that adoption actually looks like from the IT seat: who is using what, what it costs, and how much of it IT can see.
In software teams, AI usage is effectively universal.
Enterprise developers who have used AI coding tools at work, across the US, Brazil, Germany, and India.
GitHub, AI in software development survey (2024) · 2,000 developers at 1,000+ employee companies
Professional developers who use AI tools every day. Overall use or planned use hit 84% in 2025, up from 76% the year before.
Stack Overflow, Developer Survey (2025) · ~49,000 developers globally
Tech professionals using AI at work, with over 80% reporting productivity gains from it.
Google Cloud, DORA State of AI-assisted Software Development (2025) · ~5,000 tech professionals globally
The tools arrived before the policy did.
Bottom-up adoption is the defining pattern in tech companies. Engineers, support, and go-to-market teams picked their own tools and built their own workflows. The IT question is no longer whether to allow AI; it is whether you can see what is already running.
AI users who bring their own AI tools to work rather than waiting for an official rollout.
Microsoft + LinkedIn, Work Trend Index (2024) · 31,000 knowledge workers, 31 countries
Tech workers using AI at work without official approval, in a tech-heavy sample much like a SaaS company's own staff.
Retool, State of AI (2024) · 730 tech professionals
Organizations that suspect or have evidence employees use prohibited public gen-AI tools.
Gartner, GenAI blind spots survey (2025) · 302 cybersecurity leaders
Average added breach cost when shadow AI is involved. One in five breached organizations traced a breach to unsanctioned AI.
IBM + Ponemon, Cost of a Data Breach (2025) · 600 breached organizations globally
Coding tools own the budget. Visibility hasn't kept up.
2025 spend on AI coding tools, 55% of all departmental AI spend ($7.3B, up 4.1x in a year). IT operations is the next line at $700M.
Menlo Ventures, State of Generative AI in the Enterprise (2025) · 495 US enterprise AI decision-makers + market sizing
Organizations with visibility into their AI software spend, while 59% say wasted AI spend increased year over year.
Flexera, State of ITAM (2026) · 512 technology professionals
Leaders with full, real-time visibility into AI operating costs. Only 36% have direct token or usage controls.
KPMG, AI Quarterly Pulse Q2 (2026) · 204 US C-suite leaders, $1B+ revenue
Agents are already inside the stack you manage.
Organizations actively deploying AI agents, early 2024 to early 2026. Technology departments lead the deployment at 78%.
KPMG, AI Quarterly Pulse Q1 (2026) · 237 US C-suite leaders, $1B+ revenue
Enterprise apps with task-specific AI agents, 2025 versus predicted end of 2026. Agents arrive inside the software you already own, planned or not.
Gartner, enterprise apps prediction (2025) · Analyst prediction
Share of agentic AI projects Gartner predicts will be canceled by end of 2027, on escalating costs, unclear value, and weak risk controls. Choosing the right processes is the difference.
Gartner, agentic AI prediction (2025) · Analyst prediction
Use-case volume is not the same as outcomes.
Organizations that have deployed more than 100 AI use cases, versus those saying the efforts drive meaningful business outcomes.
ServiceNow + Oxford Economics, Enterprise AI Maturity Index (2025) · ~4,500 executives worldwide
Tech professionals with little or no trust in AI-generated code, alongside 90% adoption. Teams ship with tools they don't fully trust.
Google Cloud, DORA State of AI-assisted Software Development (2025) · ~5,000 tech professionals globally
Developers whose biggest AI frustration is solutions that are almost right, with 45% saying debugging AI-generated code takes longer than writing it.
Stack Overflow, Developer Survey (2025) · ~49,000 developers globally
These are the market's numbers. Get yours.
WIQ observes how work actually happens on your team and ranks what is worth automating. The same questions these surveys ask, answered with your own data.
Get a demo