Munich Startup
Chaos in administration – How Summ AI helps government agencies understand themselves

Chaos in administration – How Summ AI helps government agencies understand themselves

Kyrill Ring

Kyrill Ring

Summ AI founder Vanessa Theel talks in the video podcast Pitch & People about the blind spots of German bureaucracy and how her startup uses AI to bring order to administrative chaos. You’ll also learn why true digitalization only begins when administrations understand their own processes.

October 29, 2025

PITCH & PEOPLE Folge 8: Summ AI

Summ AI founder Vanessa Theel talks in the video podcast Pitch & People about the blind spots of German bureaucracy and how her startup uses AI to bring order to administrative chaos. You’ll also learn why true digitalization only begins when administrations understand their own processes.

DeeptechStartup

AI for clarity – outward and inward

Summ AI, founded by Flora Geske and Vanessa Theel, became known initially through a tool that translates texts into plain or simple language. This enables citizens to understand government letters and forms, even without a law degree.

“We make administration more digital and better for citizens,”

says Theel in the Pitch & People interview.

You can also translate journalistic texts from Munich Startup, such as this one, into plain language with a single click in our menu bar at the top right.

But the founders quickly realized: the problem lies not only in outward-facing language, but also in the chaotic processes within. In conversations with digitalization officers from various government agencies and administrations, it became clear:

“In 99 percent of cases, they couldn’t tell us which processes were well digitalized because they don’t even know themselves. When citizens wait 18 months for a letter, it’s not just because of the language – it’s also due to chaotic processes.”

This is exactly where the new Summ AI tool comes in. It automatically documents processes that employees describe via voice or chat. This creates transparent process models that show where work steps are duplicated, pending, or can be automated.

From an aunt to technology

The original idea goes back to the aunt of Theel’s co-founder Flora Geske, who belonged to the target audience for plain language. She had difficulty understanding complex texts – for example, from Tagesschau news.

“We thought: this is a no-brainer. AI is ideal for simplifying texts,”

recalls Vanessa Theel.

Today, Summ AI supports administrations and organizations in making information understandable and barrier-free. Some municipalities already send notices in duplicate – once in the original, once in plain language. For Theel, this is a step toward true participation:

“Administration should meet people where they are.”

Ten million people benefit directly

According to Theel, Summ AI’s core target group is around ten million people in Germany – people with learning difficulties or educational disadvantages, as well as older people with limited concentration.

“But actually, all 80 million benefit. Everyone knows situations where you’re not an expert and feel overwhelmed by technical jargon.”

Major partners like Apotheken Umschau now also rely on plain language. Over 200 articles there have already been made barrier-free – from disease descriptions to medication information. Summ AI combines rule-based procedures with AI models to simplify texts without changing content.

A startup with staying power

Summ AI operates as a B2G startup, meaning it works with administrations and social organizations. The SaaS model is classic: licenses depend on the number of users and text volume. But the market requires patience:

“You need to have a long-term perspective and high frustration tolerance. But the collaboration is incredibly valued,”

says Vanessa Theel.

Despite growing competition for AI talent – also from major players like OpenAI in Munich – motivation remains high. Theel describes the startup day-to-day as a roller coaster, but with a clear mission: reduce complexity – in language and structure.

More like this

Related Articles to Read Next

Munich Startup Recap 30/26
ecosystem

Munich Startup Recap 30/26

26.07.26
2 Min.
AI Nation launches new application round
competitions

AI Nation launches new application round

25.07.26
2 Min.
Munich Innovation Prize 2026: Four Teams Receive 10,000 Euros Each
ecosystem

Munich Innovation Prize 2026: Four Teams Receive 10,000 Euros Each

24.07.26
4 Min.
Qtas: In Search of the One Cell
Interviews

Qtas: In Search of the One Cell

24.07.26
5 Min.
The Munich Startup Festival 2026: View the Highlights Now
Inside Stories

The Munich Startup Festival 2026: View the Highlights Now

23.07.26
1 Min.
Munich Startup Festival 2026 – Recap: Europe’s Innovation Power Meets Munich Festival Flair
ecosystem

Munich Startup Festival 2026 – Recap: Europe’s Innovation Power Meets Munich Festival Flair

23.07.26
5 Min.
€4.2 Million for E-Car Platform Aampere
Deals

€4.2 Million for E-Car Platform Aampere

23.07.26
2 Min.
8pm Social wins Munich Startup Award 2026
ecosystem

8pm Social wins Munich Startup Award 2026

22.07.26
2 Min.
“Why do we stream everything – but not journalism?” How Articly came about
Podcast

“Why do we stream everything – but not journalism?” How Articly came about

22.07.26
“Why do we stream everything – except journalism?” How Articly came about
Podcast

“Why do we stream everything – except journalism?” How Articly came about

22.07.26
4 Min.
Munich Startup Festival brings Europe’s startup elite to Munich
ecosystem

Munich Startup Festival brings Europe’s startup elite to Munich

21.07.26
3 Min.
Stocard founder secures 12 million euros for AI tax firm Skalar
Deals

Stocard founder secures 12 million euros for AI tax firm Skalar

21.07.26
3 Min.