How DW uses Artificial Intelligence in journalism

How DW uses Artificial Intelligence in journalism

At DW, we use Artificial Intelligence (AI) as a tool to support journalism and create added value for audiences, not as an end in itself.

AI can help us improve workflows, accessibility, translation, data and content analysis and format adaptation.

However, our editorial standards do not change when AI is involved: Humans remain responsible for every journalistic decision and every item we publish.

We use AI only where it serves a clear editorial, public-interest or accessibility purpose. We do not use AI simply because it is new or attention-grabbing.

Trust, accuracy, and accountability remain the foundation of our journalism.

Our principles for the use of AI in journalism at DW

We use AI only when it offers a clear benefit - for example by improving accessibility, speeding up routine production steps, helping us work with large data sets, supporting translation, or enabling audiences to access our journalism more easily. AI should enhance journalism, not distract from it.

2) Editorial control always remains with journalists

We'll never publish AI-generated output without human review. A trained journalist must always make editorial decisions - adhering to the 'human-in-the-loop' principle - and DW's reporters and editors always take responsibility for the quality of our content.

3) AI should support - not replace - journalism

At DW, AI is intended to support journalistic work, not replace journalism. Quality journalism requires verification, editorial judgment, ethical consideration, contextual understanding and accountability. Those responsibilities cannot be outsourced to a machine. AI may help with repetitive or time-consuming tasks, but it does not replace reporting, investigation, editorial responsibility or the human relationship with sources and our audience.

4) Accuracy, verification and source integrity

We do not treat AI-generated output as a source of information in itself. AI systems can produce output that seems convincing but is inaccurate, incomplete or misleading. Where AI supports research or production, we always verify its claims independently, check original sources and assess whether the output meets DW’s editorial standards.

We are especially careful around any potential involvement of AI in relation to reporting breaking news, crisis and conflict coverage, investigative reporting and sensitive topics where mistakes may harm people or undermine trust.

5) Responsible use of synthetic media

Synthetic media, i.e. content that has been generated or altered by AI, can be useful in limited circumstances. Examples include accessibility, translation, format adaptation, where editorially justified and ethically appropriate. However, it can also carry a risk of impacting audience trust. We therefore apply strict standards:

Whenever we use AI in a way that audiences could reasonably consider significant – especially in the creation, presentation or distribution of content – we are transparent and clear. If AI-generated or AI-altered content could be mistaken for authentic human-created or real-world material, we clearly label it.

7) Approved, secure and lawful use

We only use tools and workflows that meet DW’s standards for security, editorial integrity, legal compliance and data protection. We do not upload highly sensitive or confidential information or editorial material into unapproved systems. When we rely on third-party tools, we assess their limitations, risks and terms of use carefully.

8) Fairness, diversity and inclusion

AI systems can reproduce or amplify biases present in training data and design choices. We therefore pay close attention to the inherent risk of stereotyping, exclusion and discrimination. At the same time, we also use AI where it can improve accessibility and inclusion – for example through translation, subtitling or more accessible formats.

9) Environmental responsibility

We recognize that training and running large AI systems can consume significant amounts of energy and resources. In our assessment of whether and how we use AI, the environmental impact is an important factor. We aim to choose proportionate solutions and use AI where the editorial or public benefit justifies its cost.

10) Accountability and continuous review

As AI technology advances quickly, we review standards continuously. We test new applications carefully, learn from pilot projects and update our practices as tools, risks, regulations and audience expectations evolve. Accountability remains with DW and its editorial leadership at all times.

Why does DW use AI in its journalism at all?

We use AI when it provides a real benefit – for example by improving accessibility, supporting translation and subtitling, writing titles and teasers, helping process large amounts of information, or making repetitive production tasks more efficient. The aim is to enhance journalism and service for audiences, not to use AI for its own sake.

Will AI replace journalists at DW?

No. AI can support certain tasks, but it cannot replace reporters, editors or producers. We do not use AI to write entire articles or scripts, or to replace human journalists in our audio or video content. Editors remain responsible for verification, judgment, ethics, context and final publication decisions.

What does "human in the loop" mean in practice?

It means that a person remains responsible at decisive points in the workflow: for selecting the use case, checking the output, correcting errors, deciding whether it is appropriate, and approving publication. AI output is never treated as self-sufficient editorial content.

How will I know if AI was used?

If AI is used in a way that is relevant to understanding the content – especially when content is synthetic, altered or highly automated – we will label it clearly. Where useful, we will also explain how and why AI was used.

Why does DW sometimes use synthetic voices?

Synthetic voices may be useful for accessibility, translation, multilingual services or other clearly justified editorial purposes. However, because voices are closely linked to authenticity and trust, we use them carefully and label them clearly.

Does DW use AI-generated images or video?

Only in limited and clearly controlled circumstances. We are especially cautious with photorealistic synthetic visuals, because audiences may interpret them as documentary evidence. When such material is used, such as reporting on AI and its capabilities, it is labelled clearly.

No. AI may assist with routine production tasks, such as summarizing, translating or adapting content formats, but we do not hand editorial responsibility to AI. Journalistic publication requires human verification, editing and approval.

How do you deal with errors, hallucinations or bias?

We assume AI systems can be wrong. That is why we verify outputs, check sources, test systems for reliability and remain alert to stereotypes, discrimination and contextual errors. Human oversight is essential precisely because AI can produce persuasive but flawed results.

What about privacy, copyright and data security?

We use tools and workflows that meet our strict legal, editorial and security requirements. Sensitive material should not be entered into unapproved systems. We also look critically at the provenance of training data and at the obligations attached to external tools and providers.

What about the environmental impact of AI?

AI can require substantial computing power and energy. We take that into account when selecting tools and use cases and weigh environmental cost against editorial and public value.

What if an external third party provides AI-generated content?

AI material generated by third parties requires careful scrutiny as to whether its use is editorially necessary. In some cases, such as when we report on AI, it is necessary to include AI images or video that has been generated by a third party. If such material is used, we identify its origin appropriately and label it as AI-generated. DW is responsible for the context in which it appears.