Valliance

Get in touch

Why Valliance

Who we are

What we do

What we think

Join us

Get in touch

What we think

News & Events

"Models that modify their own weights mid-inference destroy traceability, break safety reinforcement, and create trivial attack vectors for manipulating outputs at scale."

Self-Adapting Language Models

Dec 12, 2025

2 Mins

Themes

Trust

Topics

Thought Leadership

Leadership

Dom Selvon

Value Partner - CTO

Are you ready to shape the future enterprise?

Get in touch, and let's talk about what's next.

Get in touch

Are you ready to shape the future enterprise?

Get in touch, and let's talk about what's next.

Get in touch

Are you ready to shape the future enterprise?

Get in touch, and let's talk about what's next.

Get in touch

Valliance Newsletter

Insights and thinking direct to your inbox.

Join the Valliance Newsletter

Valliance Newsletter

Insights and thinking direct to your inbox.

Join the Valliance Newsletter

Valliance Research Report

January 2026

Discover why AI projects really stall, and how to fix them

Valliance Research Report

January 2026

Discover why AI projects really stall, and how to fix them

Valliance Research Report

January 2026

Discover why AI projects really stall, and how to fix them

Valliance Research Report

January 2026

Discover why AI projects really stall, and how to fix them

_Related thinking

View all thinking

_Related thinking

View all thinking

_Related thinking

View all thinking

_Related thinking

View all thinking

Valliance

Trust

Image description -

Valliance is now ISO 27001 certified

Dom Selvon

Feb 2, 2026

5 Mins

Valliance

Trust

Image description -

Valliance is now ISO 27001 certified

Dom Selvon

Feb 2, 2026

5 Mins

Valliance

Trust

Image description - Valliance ISO 27001 Certification

Valliance is now ISO 27001 certified

Dom Selvon

Feb 2, 2026

5 Mins

Valliance

Trust

Image description -

Valliance is now ISO 27001 certified

Dom Selvon

Feb 2, 2026

5 Mins

_Explore our themes

Let’s put AI to work.

Get in touch

London

The Hague

Join the Valliance Newsletter

Contact

Let’s put AI to work.

Get in touch

London

The Hague

Join the Valliance Newsletter

Contact

Let’s put AI to work.

Get in touch

London

The Hague

Join the Valliance Newsletter

Contact

Let’s put AI to work.

Get in touch

London

The Hague

Join the Valliance Newsletter

Contact

"Models that modify their own weights mid-inference destroy traceability, break safety reinforcement, and create trivial attack vectors for manipulating outputs at scale."

Are you ready to shape the future enterprise?

Are you ready to shape the future enterprise?

Are you ready to shape the future enterprise?

Insights and thinking direct to your inbox.

Valliance Newsletter

Discover why AI projects really stall, and how to fix them

Discover why AI projects really stall, and how to fix them

Discover why AI projects really stall, and how to fix them

Discover why AI projects really stall, and how to fix them

_Related thinking

_Related thinking

_Related thinking

_Related thinking

_Explore our themes

_Explore our themes

_Explore our themes

_Explore our themes

Ontologies

People First

Value

Trust

The Future Enterprise

Agentic Systems

Evolving Enterprise Technology

Ontologies

People First

Value

Trust

The Future Enterprise

Agentic Systems

Evolving Enterprise Technology

Valliance

Ontologies

People First

Value

Trust

The Future Enterprise

Agentic Systems

Evolving Enterprise Technology

Valliance

Ontologies

People First

Value

Trust

The Future Enterprise

Agentic Systems

Evolving Enterprise Technology

Valliance

Ontologies

People First

Value

Trust

The Future Enterprise

Agentic Systems

Evolving Enterprise Technology

Valliance

Let’s put AI to work.

Let’s put AI to work.

Let’s put AI to work.

Let’s put AI to work.