Sad news about Tony Lofthouse

[Patrick Hammer]

Hi everyone!

I'm very sad to share that Tony Lofthouse passed away this Saturday, October 25, after a long battle with cancer.

Tony's contributions to the AGI community will never be forgotten. His pioneering work on advanced control mechanisms for NARS, inspired by spiking neural networks for contextual processing, priming, and distributed computation, not only made modern NARS implementations practical but opened new directions in understanding intelligent control and adaptive systems.

Just prior to his passing, he completed his book "INSIGHT and the Unified Principle of Cognition" ( INSIGHT denoting Integrated Network for Scalable Intelligent General Higher-order Thought ), which will soon be publicly available, pending final review. It represents a major extension of his previous research, branching beyond NARS and building on his earlier work on ALANN (Adaptive Logic and Neural Network), which strongly influenced the control mechanisms and memory structures later used in OpenNARS for Applications. He also collaborated extensively on exploring Relational Frame Theory in connection with NARS, together with Robert Johansson, Robert Wünsche, Pei Wang, and me. The complexities involved in explaining and reproducing human relational learning abilities with AI (and particularly NARS) drove him to seek key underlying principles rather than simply building more complex systems.

[In memory - Stockholm AGI lab group picture @ Digital Futures, KTH & SU]

Tony leaves behind several fascinating and heavily-unexplored ideas that may hold keys to AGI, including:

- That sufficiently large interlinked concept/belief networks could exhibit emergent representational phenomena essential for robust thought patterns, ultimately supporting decision-making and purpose-driven common-sense reasoning. I remain convinced that his ALANN 2018 prototype ( https://github.com/opennars/ALANN2018 ) was a strong proof of concept for this idea and deserves further analysis (as only a simplified form of it went into ONA), and that his new book provides many insights of how to go beyond that POC.

- That studying complex dynamical systems with stochastic operating principles, particularly biological neural networks, may shed light on how distributed, self-organizing processes achieve coherent coordination in adaptive reasoning and decision making. Tony viewed cognition as an emergent control process that balances stability and plasticity through self-regulating dynamics, enabling continuous learning without catastrophic forgetting.

In research, Tony often reminded us that recognizing what remains unexplored is more important than elaborating on what we already know will not lead us closer to the goal. This adventurous spirit defined Tony, from his time as a Royal Marine to his life as an independent researcher who left a stable career to pursue his dreams and sail the oceans.
His vision, intellect, and kindness profoundly touched everyone who had the privilege to work with him. He will be missed, but his ideas will continue to inspire future work, and a part of him will live on in all of us.

Lastly, we have a mission to fulfill, a project we promised him to complete, Project 2029. Tony believed that the core principles are already out there, waiting to be combined. Regardless of personal belief, let’s push harder toward the kind of adaptive systems we truly want to see, the kind still missing from today's AI.

Thoughtful regards,
Patrick

[Pei Wang]

Tony will always be with us!

Pei

Virus-free.www.avast.com

--
You received this message because you are subscribed to the Google Groups "open-nars" group.
To unsubscribe from this group and stop receiving emails from it, send an email to open-nars+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/open-nars/8c08c98c-8a54-4188-bf07-e4136a3c5b8en%40googlegroups.com.

[Dwanev]

Hi Patrick,

Thank you for sharing this news with us. I think I speak for others as well when I say I’ll miss him.

Our thoughts go to his family at this time.

Kind regards 

Dwane

On 28 Oct 2025, at 23:08, Patrick Hammer <pat...@gmail.com> wrote:

Hi everyone!

I'm very sad to share that Tony Lofthouse passed away this Saturday, October 25, after a long battle with cancer.

Tony's contributions to the AGI community will never be forgotten. His pioneering work on advanced control mechanisms for NARS, inspired by spiking neural networks for contextual processing, priming, and distributed computation, not only made modern NARS implementations practical but opened new directions in understanding intelligent control and adaptive systems.

Just prior to his passing, he completed his book "INSIGHT and the Unified Principle of Cognition" ( INSIGHT denoting Integrated Network for Scalable Intelligent General Higher-order Thought ), which will soon be publicly available, pending final review. It represents a major extension of his previous research, branching beyond NARS and building on his earlier work on ALANN (Adaptive Logic and Neural Network), which strongly influenced the control mechanisms and memory structures later used in OpenNARS for Applications. He also collaborated extensively on exploring Relational Frame Theory in connection with NARS, together with Robert Johansson, Robert Wünsche, Pei Wang, and me. The complexities involved in explaining and reproducing human relational learning abilities with AI (and particularly NARS) drove him to seek key underlying principles rather than simply building more complex systems.

<team.jpg>

[In memory - Stockholm AGI lab group picture @ Digital Futures, KTH & SU]

Tony leaves behind several fascinating and heavily-unexplored ideas that may hold keys to AGI, including:

- That sufficiently large interlinked concept/belief networks could exhibit emergent representational phenomena essential for robust thought patterns, ultimately supporting decision-making and purpose-driven common-sense reasoning. I remain convinced that his ALANN 2018 prototype ( https://github.com/opennars/ALANN2018 ) was a strong proof of concept for this idea and deserves further analysis (as only a simplified form of it went into ONA), and that his new book provides many insights of how to go beyond that POC.

- That studying complex dynamical systems with stochastic operating principles, particularly biological neural networks, may shed light on how distributed, self-organizing processes achieve coherent coordination in adaptive reasoning and decision making. Tony viewed cognition as an emergent control process that balances stability and plasticity through self-regulating dynamics, enabling continuous learning without catastrophic forgetting.

In research, Tony often reminded us that recognizing what remains unexplored is more important than elaborating on what we already know will not lead us closer to the goal. This adventurous spirit defined Tony, from his time as a Royal Marine to his life as an independent researcher who left a stable career to pursue his dreams and sail the oceans.
His vision, intellect, and kindness profoundly touched everyone who had the privilege to work with him. He will be missed, but his ideas will continue to inspire future work, and a part of him will live on in all of us.

Lastly, we have a mission to fulfill, a project we promised him to complete, Project 2029. Tony believed that the core principles are already out there, waiting to be combined. Regardless of personal belief, let’s push harder toward the kind of adaptive systems we truly want to see, the kind still missing from today's AI.

Thoughtful regards,
Patrick

--

You received this message because you are subscribed to the Google Groups "open-nars" group.
To unsubscribe from this group and stop receiving emails from it, send an email to open-nars+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/open-nars/8c08c98c-8a54-4188-bf07-e4136a3c5b8en%40googlegroups.com.

<team.jpg>

["K. R. Thórisson"]

I am very saddened to hear this news. 

I did not know of his illness.

=K

On 28 Oct 2025, at 23:08, Patrick Hammer <pat...@gmail.com> wrote:

Hi everyone!

I'm very sad to share that Tony Lofthouse passed away this Saturday, October 25, after a long battle with cancer.

Tony's contributions to the AGI community will never be forgotten. His pioneering work on advanced control mechanisms for NARS, inspired by spiking neural networks for contextual processing, priming, and distributed computation, not only made modern NARS implementations practical but opened new directions in understanding intelligent control and adaptive systems.

Just prior to his passing, he completed his book "INSIGHT and the Unified Principle of Cognition" ( INSIGHT denoting Integrated Network for Scalable Intelligent General Higher-order Thought ), which will soon be publicly available, pending final review. It represents a major extension of his previous research, branching beyond NARS and building on his earlier work on ALANN (Adaptive Logic and Neural Network), which strongly influenced the control mechanisms and memory structures later used in OpenNARS for Applications. He also collaborated extensively on exploring Relational Frame Theory in connection with NARS, together with Robert Johansson, Robert Wünsche, Pei Wang, and me. The complexities involved in explaining and reproducing human relational learning abilities with AI (and particularly NARS) drove him to seek key underlying principles rather than simply building more complex systems.

<team.jpg>

[In memory - Stockholm AGI lab group picture @ Digital Futures, KTH & SU]

Tony leaves behind several fascinating and heavily-unexplored ideas that may hold keys to AGI, including:

- That sufficiently large interlinked concept/belief networks could exhibit emergent representational phenomena essential for robust thought patterns, ultimately supporting decision-making and purpose-driven common-sense reasoning. I remain convinced that his ALANN 2018 prototype ( https://github.com/opennars/ALANN2018 ) was a strong proof of concept for this idea and deserves further analysis (as only a simplified form of it went into ONA), and that his new book provides many insights of how to go beyond that POC.

- That studying complex dynamical systems with stochastic operating principles, particularly biological neural networks, may shed light on how distributed, self-organizing processes achieve coherent coordination in adaptive reasoning and decision making. Tony viewed cognition as an emergent control process that balances stability and plasticity through self-regulating dynamics, enabling continuous learning without catastrophic forgetting.

In research, Tony often reminded us that recognizing what remains unexplored is more important than elaborating on what we already know will not lead us closer to the goal. This adventurous spirit defined Tony, from his time as a Royal Marine to his life as an independent researcher who left a stable career to pursue his dreams and sail the oceans.
His vision, intellect, and kindness profoundly touched everyone who had the privilege to work with him. He will be missed, but his ideas will continue to inspire future work, and a part of him will live on in all of us.

Lastly, we have a mission to fulfill, a project we promised him to complete, Project 2029. Tony believed that the core principles are already out there, waiting to be combined. Regardless of personal belief, let’s push harder toward the kind of adaptive systems we truly want to see, the kind still missing from today's AI.

Thoughtful regards,
Patrick

--