BearingPoint: DACH codes with AI, but barely embeds it
This article is an AI-generated translation of the German original. The German version is authoritative.
3 Min. Read Time
91 percent use AI in software development. 18 percent have embedded it into their processes. In between lies reality: 73 percent rely on AI sporadically.
Key Takeaways
- 91 percent use AI in development. 18 percent have integrated the tools into workflows, approvals, and architecture.
- 73 percent stick to ad-hoc usage. No surveyed company rules out AI entirely.
- 70 percent cite security and compliance risks as barriers. Three out of four companies have internal AI guidelines, either fully or partially.
- 59 percent aim to expand AI expertise. Equal shares (27 percent each) plan to grow junior and senior teams.
Related:Skills Shortage: Logistics Turns to AI Instead of Training / McKinsey: AI Speeds Up Operations, But EBIT Remains Flat
What does “fully integrated” mean? The tools are embedded in workflows, approvals, and architecture. Ad-hoc usage remains a secondary approach.
91 to 18: The Rest Still Experiment
The number every executive knows is the first: nearly all companies deploy AI in software development. The number that defines the process is the second: almost no one has embedded it where code is approved and architecture is decided.
73 percent use the tools only sporadically. No company in the sample excludes AI outright. AI has arrived, but it’s not yet industrialized.
BearingPoint surveyed 112 executives across Germany, Austria, and Switzerland. 46 percent were from Germany, with the rest split between Austria and Switzerland.
Matthias Roeser, global head of Technology at BearingPoint, states: “With AI, nearly every project becomes a software project, and after countless lab experiments, AI has arrived in corporate software development, but it’s not yet industrialized.” His words capture the 18 percent. He also highlights productivity gains.
Documentation and Code First, Processes Later
Where AI is applied, the focus is tight. Documentation (77 percent), code generation (76 percent), and quality assurance/code reviews (74 percent) are the top areas where text and patterns emerge. These aren’t the stages where companies overhaul workflows.
| Development Step | Percentage of Respondents |
|---|---|
| Documentation | 77 percent |
| Code Generation | 76 percent |
| Quality Assurance and Code Reviews | 74 percent |
Data from BearingPoint, 112 executives surveyed, April 2026.
81 percent report faster development. 77 percent see advantages in building initial prototypes. 74 percent experience relief from routine tasks. 59 percent observe improved overall efficiency.
Security, not lack of will, is the real brake on AI adoption
70 percent cite security and compliance risks as barriers. Tool selection follows at 61 percent, while integration into existing processes stands at 58 percent. Three out of four companies have introduced or partially implemented internal AI guidelines. Rules exist. AI has a fixed place in the process for only 18 percent.
Yet expansion is on the horizon. 59 percent plan to boost AI expertise, with 27 percent each aiming to grow both junior and senior teams.
Stefan Zimmermann, Partner at BearingPoint, sees the next step in organizational integration: AI is here, but the process is still catching up.
Frequently Asked Questions
Who did BearingPoint survey?
112 executives from Germany, Austria, and Switzerland, surveyed online in April 2026. 46 percent were from Germany, while Austria and Switzerland each contributed just over a quarter.
What does “fully integrated” mean?
The tools are permanently embedded in workflows, approvals, and architecture. Ad-hoc use remains alongside it. Only 18 percent fall into the fully integrated category, while 73 percent do not.
What is the staffing outlook?
59 percent plan to expand AI expertise, with 27 percent each aiming to grow junior and senior teams.
Why do so few industrialize AI?
70 percent point to security and compliance risks. Tool selection is a concern for 61 percent, while 58 percent struggle with integration into existing processes.
Editor’s Picks
- Skills Shortage: Logistics turns to AI instead of training
- Gartner: 85 percent plan higher AI budgets
- McKinsey: AI speeds up operations, but EBIT stays flat
Read more on MyBusinessFuture
MyBusinessFutureB2B platforms: Why marketplaces are changing distributionMyBusinessFutureInflation falls: Tax margins instead of hopingMyBusinessFutureNIS-2 Registration: What the Executive Suite Must Do NowMore from the MBF Media Network
cloudmagazinTerraform Apply in the Cloud: Command from the USA?Digital ChiefsWhy AI fails at master dataSecurityTodayExploitGym: OpenAI agents built a secret channelImage source: AI-generated (September 2026)
Translated from the German original with AI support. The German version is authoritative.
