AI Strategies and Quantum Computing Insights for 2026

Introduction

The roadmap for 2026 includes the following topics:

• AI Strategy
• Agentic Software Development
• Quantum Computing and Quantum Machine Learning (QML)
• Cryptography and Post-Quantum Cryptography (PQC)

The sections below describe each topic in detail.

AI Strategy

Are we now dealing with an AI hype or is it an LLM hype?

Some believe that the current AI hype is actually a hype of Large Language Models (LLM).

AI is not just a hype. It has already settled in our way of living. It is also a permanent part of our working life. But some experts say that the current Multimodal LLMs (MLLM) will never be able to replace the human brain. They also believe the massive investments in these developments will cease at some point and the bubble will burst.

A new phenomenon is popping up, World Model, a neural network that understands the dynamics of the real world. It uses input data, such as text, image, video and movement to generate videos that simulate realistic physical environments.

We also follow the development of AI agents, agent loops and multi-agent architectures.

Agentic Software Development

Today, 80% of the software developers use Agentic Coding tools. Examples include Claude Code of Anthropic and Devstral of Mistral AI. These 2 tools are investigated in more detail and some prototypes will be built.

These tools are not only for developing and debugging new code, but will also assist in mitigating legacy code. Many tech companies are currently maintaining large amounts of code developed in the 80’s and 90’s of the previous century. This code is end-of-life and should be replaced.

Quantum Computing and Quantum Machine Learning (QML)

In a previous post, the current developments of quantum computing are outlined and their importance for AI. Current machine learning algorithms are tailored to run on quantum machines. We will investigate the impact on the development of current (electronic) AI chips. Will the latter still be necessary? Or is the massive computing power needed to train AI models reduced to a nutshell?

Cryptography and Post-Quantum Cryptography

Asymmetric cryptography used for secured communication and digital signatures in Blockchain are discussed. Quantum computing poses a threat to current algorithms, such as RSA and ECDSA. Post-Quantum algorithms are being developed to mitigate these problems.

Summary

In 2026, we will focus on AI strategy, agentic coding, quantum computing and cryptography. The goal is to advise companies on these topics.

Glossary of Terms

• Multimodal LLM – Type of LLM designed to process and reason across multiple data modalities such as text, images and audio
• Agentic coding – An approach in software development where AI acts as a collaborative partner that assists the developer in writing, refactoring and debugging code
• Quantum machine learning – Refers to quantum algorithms that solve tasks in machine learning, thereby improving and often expediting classical machine learning techniques
• Asymmetric cryptography – Known as public-key cryptography, which uses two separate keys: one key (public) is used to encrypt or sign the plain text, and the second key (private) is used to de-crypt the cipher text or verify the signature
• RSA – Algorithm developed by Ron Rivest, Adi Shamir and Leonard Adleman
• ECDSA – Elliptic Curve Digital Signature Algorithm