Karl Friston

Karl John Friston was born on July 12, 1959, in York, UK. He is married with three children.⁠[1]

He received a BA from Gonville and Caius College, Cambridge, in 1980 (physics and psychology) and an MBBS from King’s College Medical School in 1983, trained in psychiatry in Oxford from 1985 to 1988, and obtained the MRCPsych in 1988.⁠[1]

From 1988 to 1992 he worked at the MRC Cyclotron Unit, Hammersmith Hospital, London, and from 1992 to 1994 he was a fellow in theoretical neurobiology at the Neurosciences Institute in La Jolla, California.⁠[2]

The methodology was described in two papers in 1990 and 1991, the first by Friston et al. in 1990; it was encoded in Matlab and used extensively by the MRC Cyclotron Unit at Hammersmith Hospital.⁠[3]

The SPM software suite was originally developed by Friston at the MRC Cyclotron Unit for statistical analysis of positron emission tomography (PET) functional imaging data and was made available to the functional imaging community in 1991; in the summer of 1994 he wrote most of SPM’94, its first major revision.⁠[4]

He became a Wellcome Senior Research Fellow at the Institute of Neurology in 1994, a professor at the UCL Institute of Neurology in 1998 and, from 2001, Scientific Director of the Wellcome Trust Centre for Neuroimaging.⁠[2]

In 1995 Friston and Chris Frith proposed the dysconnection hypothesis of schizophrenia (alongside Weinberger’s 1993 work), locating its pathology in a failure of functional integration in the brain; in 2016 he and co-authors reviewed the hypothesis twenty years on.⁠[5]

He received the first Young Investigators Award in Human Brain Mapping in 1996, was elected a Fellow of the Academy of Medical Sciences in 1999 and served as President of the international Organization of Human Brain Mapping in 2000.⁠[6]

In 2000 he and John Ashburner published “Voxel-Based Morphometry—The Methods” in NeuroImage, describing a voxel-wise comparison of local gray-matter concentration between two groups of subjects.⁠[7]

In 2003 he, L. Harrison and W. Penny published “Dynamic causal modelling” in NeuroImage, presenting a Bayesian approach to identifying nonlinear input–state–output systems for the analysis of effective connectivity in fMRI.⁠[8]

He received the Minerva Golden Brain Award in 2003, was elected a Fellow of the Royal Society in 2006, received a Collège de France medal in 2008, an honorary doctorate from the University of York in 2011 and, in 2013, the Weldon Memorial Prize and Medal for contributions to mathematical biology.⁠[9]

In January 2010 he published “The free-energy principle: a unified brain theory?” in Nature Reviews Neuroscience, examining brain theories such as neural Darwinism, information theory and optimal control from the free-energy perspective and arguing that each optimizes the same quantity, value or its complement, surprise (prediction error).⁠[10]

He was elected to EMBO in 2014 and to the Academia Europaea in 2015; he also holds honorary doctorates from the universities of York, Zurich, Liège and Radboud.⁠[11]

In 2016 he won the Charles L. Branch BrainHealth Award, given each year by the Center for BrainHealth to a neuroscientist whose innovation has greatly contributed to brain research, and gave the keynote at the center’s research symposium on April 14.⁠[12]

In 2016 the Organization for Human Brain Mapping gave him the Glass Brain Award for lifetime accomplishments. Cathy Price, director of the Centre for Neuroimaging, said his SPM framework underpinned the massive acceleration in imaging and that his free energy principle was engaging philosophers, robot engineers and theoretical physicists.⁠[13]

He appeared on episode 99 of the Lex FridmanLex FridmanLex Fridman is a Soviet-born, US-raised computer scientist and podcast host who works as an MIT research scientist on human-AI interaction and autonomous driving. His Lex Fridman Podcast is known for hours-long conversations with guests including heads of state and technology leaders.Open the full entry Podcast, whose description says he has been cited over 245,000 times and is known for ideas in brain imaging, theoretical neurobiology and the free-energy principle; topics included how much of the brain we understand, brain imaging and its history, Neuralink and brain-computer interfaces, the free energy principle and the meaning of life.⁠[14]

In 2022 the MIT Press published Active Inference: The Free Energy Principle in Mind, Brain, and Behavior, co-written with Thomas Parr and Giovanni Pezzulo and described as the first comprehensive treatment of active inference, which characterizes perception, planning and action as probabilistic inference.⁠[15]

On December 1, 2022, VERSES Technologies announced that Friston had joined as Chief Scientist to lead its advanced AI research and development; the first planned output of his tenure was a white paper setting out a multi-stage roadmap toward artificial general intelligence.⁠[16]

In December 2024 VERSES AI said it had extended its exclusive contract with Friston, who continues as Chief Scientist, with performance-based incentives tied to his contributions to its main product, Genius, among other milestones.⁠[17]

On October 24, 2025, he gave a talk, “The physics of sentience,” at the Pattern Recognition Lab of Friedrich-Alexander-Universität Erlangen-Nürnberg in Germany.⁠[6]

Personal website: Public link: https://www.fil.ion.ucl.ac.uk/~karl/ (opens in a new window)⁠[14]

University College London: Official website: https://www.ucl.ac.uk/ (opens in a new window)⁠[13]