Generated by All in One SEO v4.9.5, this is an llms.txt file, used by LLMs to index the site. # Safe Intelligence Machine Learning Model Verification & Robustification ## Sitemaps - [XML Sitemap](http://dev-safeintelligence.pantheonsite.io/sitemap.xml): Contains all public & indexable URLs for this website. ## Posts - [Resources Hub](http://dev-safeintelligence.pantheonsite.io/resources-hub/) - [Dynamic Back-Substitution in Bound-Propagation-Based Neural Network Verification](http://dev-safeintelligence.pantheonsite.io/dynamic-back-substitution-in-bound-propagation-based-neural-network-verification/) - Kouvaros, P., Brueckner, B., Henriksen, P., Lomuscio, A. (2025), Proceedings of the 39th AAAI Conference on Artificial Intelligence (AAAI25) Outcome Value The paper shows advances of state-of-the-art neural network verification by accelerating an algorithm which is used by most verification toolkits. This allows verifiers to scale to even large neural networks, aiding the certification of - [Safe Intelligence in Zagreb: Our Highlights from the 2025 Symposium on AI Verification (SAIV)](http://dev-safeintelligence.pantheonsite.io/safe-intelligence-in-zagreb-our-highlights-from-the-2025-symposium-on-ai-verification-saiv/) - Brueckner, B., Kouvaros, P., Highlights from SAIV 2025 How can we truly trust an AI system? As models learn from data in ways that aren't always transparent, ensuring they are safe and reliable, especially in critical applications, is one of the most important challenges in technology today. This is the central question that drives the - [Verification of Neural Networks Against Convolutional Perturbations via Parameterised Kernels](http://dev-safeintelligence.pantheonsite.io/verification-of-neural-networks-against-convolutional-perturbations-via-parameterised-kernels-brueckner-b-lomuscio-a-2025-proceedings-of-the-39th-aaai-conference-on-artificial-intelligence-aaa/) - Brueckner, B., Lomuscio, A. (2025), Proceedings of the 39th AAAI Conference on Artificial Intelligence (AAAI25) Outcome Value Ensuring the robustness of neural networks against real-world perturbations is critical for their safe deployment. Existing verification techniques struggle to efficiently handle convolutional perturbations due to loose bounding techniques and high-dimensional encodings. Our work advances the state-of-the-art by - [My First 30 Days as a Developer Evangelist at Safe Intelligence](http://dev-safeintelligence.pantheonsite.io/my-first-30-days-as-a-developer-evangelist-at-safe-intelligence/) - Hello world, I’m Brain (yes 🧠) After six years of technical writing and public speaking in the AI space plus experience working in the financial industry as a data scientist, I have stepped into a new role as a Developer Evangelist at Safe Intelligence. I am here to help you build and deploy better, provably - [AI Quality and Safety at the AI Engineer Summit](http://dev-safeintelligence.pantheonsite.io/ai-quality-and-safety-at-the-ai-engineer-summit/) - Every new wave of technology brings a new wave of conferences, summits, and meetups. AI is no different, and a dizzying array of new events has sprung up. One of the most interesting of these is the AI Engineer Summit series (and its counterpart, the AI Worlds Fair). The events resonate with our work at Safe Intelligence because - [Scaling AI through more powerful validation and robustness](http://dev-safeintelligence.pantheonsite.io/scaling-ai-through-more-powerful-validation-and-robustness/) - AI is fast becoming one of the building blocks of society and holds huge potential for productivity gains, automation and improvements to life. As techniques advance we make better models and enable new use-cases. However it remains much harder to validate that AI systems will act appropriately when deployed in the real world. Safe Intelligence - [Safe Intelligence Raises £4M To Deliver Advanced Validation for Reliable AI](http://dev-safeintelligence.pantheonsite.io/safe-intelligence-raises-4m-to-deliver-advanced-validation-for-reliable-ai/) - Technology offers more effective analysis of machine learning models than statistical testing, delivering safer, more resilient models; opens early access program London, UK – Feb. 20, 2025 – Safe Intelligence, a pioneer in deep validation artificial intelligence (AI) systems, today announced that it has secured £4.15M in seed funding in an investment round led by ## Pages - [Home](http://dev-safeintelligence.pantheonsite.io/) - [About](http://dev-safeintelligence.pantheonsite.io/about/) - [Product](http://dev-safeintelligence.pantheonsite.io/product/) - [Applications](http://dev-safeintelligence.pantheonsite.io/applications/) - [Contact](http://dev-safeintelligence.pantheonsite.io/contact/) - [Terms & Conditions](http://dev-safeintelligence.pantheonsite.io/terms-conditions/) - AGREEMENT TO OUR LEGAL TERMS We are Safe Intelligence Limited (‘Company’, ‘we’, ‘us’, or ‘our’), a company registered in the United Kingdom at Demsa Accounts, 278 Langham Road, London, England N15 3NP. We operate the website safeintelligence.ai (the ‘Site’), as well as any other related products and services that refer or link to these legal - [Privacy Policy](http://dev-safeintelligence.pantheonsite.io/privacy-policy/) - This privacy notice for Safe Intelligence Limited (‘we’, ‘us’, or ‘our’), describes how and why we might collect, store, use, and/or share (‘process’) your information when you use our services (‘Services’), such as when you: Visit our website at https://www.safeintelligence.ai, or any website of ours that links to this privacy noticeEngage with us in other - [FAQs](http://dev-safeintelligence.pantheonsite.io/faqs/) - Awaiting content, is this page still needed? ## Categories - [Blogs](http://dev-safeintelligence.pantheonsite.io/category/blogs/) - [News](http://dev-safeintelligence.pantheonsite.io/category/news/) - [Research](http://dev-safeintelligence.pantheonsite.io/category/research/) ## Tags - [Research](http://dev-safeintelligence.pantheonsite.io/tag/research/) - [Formal Verification](http://dev-safeintelligence.pantheonsite.io/tag/formal-verification/) - [Neural networks](http://dev-safeintelligence.pantheonsite.io/tag/neural-networks/) - [Robustification](http://dev-safeintelligence.pantheonsite.io/tag/robustification/) - [Events](http://dev-safeintelligence.pantheonsite.io/tag/events/) - [Safe Intelligence News](http://dev-safeintelligence.pantheonsite.io/tag/safe-intelligence-news/) - [Verification](http://dev-safeintelligence.pantheonsite.io/tag/verification/) - [Verifier](http://dev-safeintelligence.pantheonsite.io/tag/verifier/) - [SAIV](http://dev-safeintelligence.pantheonsite.io/tag/saiv/)