Large Language Models in Publishing

Generative AI is having a fundamental impact on sectors with large amounts of textual data. The publishing industry has much to gain from the advanced data processing capabilities of large language models (LLMs).

  • SOC 2 Compliant
  • GDPR Compliant
  • CSA STAR Level 1 Certified
  • Improve customer engagement

  • Generate new revenue streams

  • Enhance the subscriber experience

  • Optimize existing content for greater insights

  • Accelerate content creation and internal productivity

  • Build an AI muscle

Zeit Online Customer Story

Learn how a major German news site used LLMs to improve content discovery for its subscribers.

Explore Our Resources

deepset Cloud Top Use Cases for Publishing

  • Intelligent Document Processing

    Intelligent Document Processing (IDP) handles large textual datasets to extract or summarize information. deepset Cloud's IDP workflow with LLMs connects to your data to help you find insights from multiple sources. This makes it easier to process data and gain insights quickly, freeing up analysts for tasks that require their expertise.

  • Retrieval Augmented Generation

    Retrieval augmented generation (RAG) combines LLM creativity with data-driven reliability. Extend your basic RAG setup in deepset Cloud to build tools and products tailored to your use case. Our AI engineers can help you customize your pipelines and improve your prompt engineering skills.

  • Semantic Search

    Embedding models understand the meaning of documents, so they can help you find what you're looking for in a database based on semantics. The modular paradigm in deepset Cloud lets you customize search interfaces with many ranking and filtering options. You can also use embedding models to recommend similar content.

  • Text2SQL

    deepset has been a pioneer in applying LLMs to complex business intelligence (BI) tasks. The text2SQL template in deepset Cloud uses LLMs to search and interpret tabular BI databases using natural language. This allows you to compare data and gain insights across datasets.

Building with LLMs for Publishing?

We've helped many publishing companies build AI-powered products and workflow tools from scratch. Contact us to see how deepset Cloud can help you quickly ship a working prototype.

Frequently Asked Questions

  • While there are companies who have gone the route of training their own LLM from scratch - spending millions of dollars in the process - it is actually a better idea to remain flexible if you want to stay competitive. Vendor agnosticism, a principle championed by deepset, allows you to change models when they no longer serve you. For example, if a cheaper or faster model comes along, you can simply plug it into your existing pipeline and move on.

  • Keping data safe, especially sensitive customer or business data, is a big concern in the age of AI models and decentralized computing infrastructures. At deepset, we recognize this and have therefore prioritized data security. Users can manage access with MFA and SSO in deepset Cloud, while our virtual private cloud (VPC) option provides the flexibility to leave their data layer in their preferred location. Furthermore, we are SOC 2 and GDPR-compliant, as well as CSA STAR Level 1 certified.

  • To build products or internal tools with AI, you need to put together a team that understands AI technology, has a product mindset, and understands both user needs and business requirements. This type of team is called an "AI team. It can be large or small, as long as it has the right cross-functional skills. To learn more about AI teams and how to build one, check out our resource pages.

  • LLMs "hallucinate", that is, they make up facts that are not supported by any data. Because of the eloquence of these models, hallucinations can be difficult to detect, creating a volatile factor that is a barrier to using LLMs in production. However, using a combination of prevention techniques, teams can reduce the number of hallucinations to a minimum. These include effective prompting, grounding the LLM's response in fact-checked data through RAG, and monitoring the Groundedness of responses.

Contact Us to Speak to Our Experts

Don't fall behind the AI curve. Start shipping LLM-powered applications in days, not weeks.