Increase efficiency, maximize output

The use of artificial intelligence has become indispensable in many areas of business, but it has so far only been used to a limited extent in tender management. One common reason for this is that companies have concerns about data protection and the lack of transparency in data processing and storage.

Challenges in the use of AI in companies

Companies should rightly be wary of publicly accessible AI applications such as ChatGPT, as data entered or uploaded is stored outside the company infrastructure. In many cases, this does not comply with compliance guidelines. Another problem is the so-called "hallucinations". Large Language Models (LLMs), the underlying AI technology from ChatGPT, generate new content from the training data based on probabilities and statistical patterns without actually checking the content. This sometimes leads to plausible-sounding but incorrect or outdated information and encourages wrong decisions - which can have fatal consequences, especially when answering tender questions.

Despite these challenges, there is a safe use of generative AI in the corporate environment, namely through "Retrieval Augmented Generation" with an "Insight Engine".

Retrieval Augmented Generation with Insight Engine as a solution

Retrieval Augmented Generation (RAG) allows companies to link Large Language Models (LLMs) with their own data sources. The Insight Engine takes on the task of extracting relevant data from internal sources and forwarding it to the LLM together with the source information. The data remains in its original storage location and is not passed on to third parties. The LLM uses this data to formulate precise answers in natural language, which can be traced back to the relevant sources in order to check them if necessary and minimize hallucinations.

Retrieval Augmented Generation in the tendering process

The RAG approach with an insight engine enables bid managers to generate tender responses in a time-saving manner. The insight engine extracts relevant information from past tenders, RFP documents and other internal sources and passes it on to the Large Language Model (LLM). In this way, complex questions are answered in the shortest possible time - without the need for time-consuming research into the required information. This gives bid managers time to concentrate on strategic tasks instead of sifting through mountains of documents.

However, insight engines not only play a key role in the RAG approach, they also relieve bid managers right from the start of the tendering process.

Monitoring of tender portals and creation of requirement catalogs

Insight Engines search tender portals, such as the European Tenderpotal TED, to identify relevant tenders for the company. The tool compares the data from past tenders and documented RFPs in order to suggest only those tenders that are really relevant for the company.

The Insight Engine is also able to extract the requirements of tenders. The LLM uses this to formulate a requirements catalog and creates recommendations for action for bid managers.

Better bid/no-bid decision through AI-generated 360-degree views

Another strength of Insight Engines is the AI-generated 360-degree view of relevant data. This view integrates information from the company's various data sources. For example, news feeds on market movements or competitor analyses can be integrated into 360-degree views via web connectors. This enables companies to better assess whether it makes economic sense to participate in tenders or not.

It is also possible to chat with the 360-degree view. Users can ask questions quickly and easily in natural language and interact directly with the objects, such as PDFs.

The use of artificial intelligence in tender management therefore offers companies decisive advantages: Retrieval Augmented Generation (RAG) with an insight engine automates time-consuming processes, protects sensitive data and minimizes wrong decisions due to hallucinations. Insight engines also enable the targeted selection of relevant tenders, the creation of response catalogs and more informed bid/no-bid decisions. As a result, companies not only gain efficiency, but also strategic flexibility. By connecting their own data sources to LLMs and AI-supported 360-degree views, competitiveness is sustainably increased. AI in supply management can therefore be a decisive lever for success.

Mario Matuschek, AI Solution Manager at Mindbreeze