Kjell Jørgen Hole
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Kjell Jørgen Hole has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
André N. Klingsheim
Department of Informatics, University of Bergen
1 shared publicationChristian Otterstad
Simula UiB
1 shared publicationHallvar Helleseth
Avenir AS
1 shared publicationJan B. Henriksen
Department of Informatics, University of Bergen
1 shared publicationLars-Helge Netland
Department of Informatics, University of Bergen
1 shared publication- Per Kristian Eide
Institute of Clinical Medicine, Faculty of Medicine, University of Oslo, Oslo, Norway
1 shared publication
Latest Publications
- 29 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computer
- Vol. 58 (2) , 16-24
Intelligent Assistants Strengthening Personhood
The article considers future intelligent personal assistants with artificial intelligence based on the neocortex, the intelligence center of the brain. It evaluates the assistants’ effects on users and recommends measures to strengthen users’ personhood.
- 29 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computer
- Vol. 58 (2) , 16-24
Intelligent Assistants Strengthening Personhood
The article considers future intelligent personal assistants with artificial intelligence based on the neocortex, the intelligence center of the brain. It evaluates the assistants’ effects on users and recommends measures to strengthen users’ personhood.
- Open Access
- 26 May 2024
- Published by Springer Nature in Minds and Machines
- Vol. 34 (2) , 1-14
Tool-Augmented Human Creativity
Creativity is the hallmark of human intelligence. Roli et al. (Frontiers in Ecology and Evolution 9:806283, 2022) state that algorithms cannot achieve human creativity. This paper analyzes cooperation between humans and intelligent algorithmic tools to compensate for algorithms’ limited creativity. The intelligent tools have functionality from the neocortex, the brain’s center for learning, reasoning, planning, and language. The analysis provides four key insights about human-tool cooperation to solve challenging problems. First, no neocortex-based tool without feelings can achieve human creativity. Second, an interactive tool exploring users’ feeling-guided creativity enhances the ability to solve complex problems. Third, user-led abductive reasoning incorporating human creativity is essential to human-tool cooperative problem-solving. Fourth, although stakeholders must take moral responsibility for the adverse impact of tool answers, it is still essential to teach tools moral values to generate trustworthy answers. The analysis concludes that the scientific community should create neocortex-based tools to augment human creativity and enhance problem-solving rather than creating autonomous algorithmic entities with independent but less creative problem-solving.
- 26 May 2024
- Published by Springer Nature in Minds and Machines
- Vol. 34 (2) , 1-14
Open AccessTool-Augmented Human Creativity
Creativity is the hallmark of human intelligence. Roli et al. (Frontiers in Ecology and Evolution 9:806283, 2022) state that algorithms cannot achieve human creativity. This paper analyzes cooperation between humans and intelligent algorithmic tools to compensate for algorithms’ limited creativity. The intelligent tools have functionality from the neocortex, the brain’s center for learning, reasoning, planning, and language. The analysis provides four key insights about human-tool cooperation to solve challenging problems. First, no neocortex-based tool without feelings can achieve human creativity. Second, an interactive tool exploring users’ feeling-guided creativity enhances the ability to solve complex problems. Third, user-led abductive reasoning incorporating human creativity is essential to human-tool cooperative problem-solving. Fourth, although stakeholders must take moral responsibility for the adverse impact of tool answers, it is still essential to teach tools moral values to generate trustworthy answers. The analysis concludes that the scientific community should create neocortex-based tools to augment human creativity and enhance problem-solving rather than creating autonomous algorithmic entities with independent but less creative problem-solving.
- Open Access
- 7 March 2023
- Published by Springer Nature in AI and Ethics
- Vol. 4 (2) , 345-352
Tools with general AI and no existential risk
According to philosophers and scientists in artificial intelligence (AI), future autonomous agents with general AI constitute an existential risk to humanity. This paper leverages results from neuroscience to propose tools with general AI and no existential risk. Tools answering questions in different domains enable the safe exploration of general AI’s enormous potential.
- 7 March 2023
- Published by Springer Nature in AI and Ethics
- Vol. 4 (2) , 345-352
Open AccessTools with general AI and no existential risk
According to philosophers and scientists in artificial intelligence (AI), future autonomous agents with general AI constitute an existential risk to humanity. This paper leverages results from neuroscience to propose tools with general AI and no existential risk. Tools answering questions in different domains enable the safe exploration of general AI’s enormous potential.
- Open Access
- 20 July 2021
- Published by Springer Nature in SN Applied Sciences
- Vol. 3 (8) , 1-14
A thousand brains: toward biologically constrained AI
This paper reviews the state of artificial intelligence (AI) and the quest to create general AI with human-like cognitive capabilities. Although existing AI methods have produced powerful applications that outperform humans in specific bounded domains, these techniques have fundamental limitations that hinder the creation of general intelligent systems. In parallel, over the last few decades, an explosion of experimental techniques in neuroscience has significantly increased our understanding of the human brain. This review argues that improvements in current AI using mathematical or logical techniques are unlikely to lead to general AI. Instead, the AI community should incorporate neuroscience discoveries about the neocortex, the human brain’s center of intelligence. The article explains the limitations of current AI techniques. It then focuses on the biologically constrainedThousand Brains Theorydescribing the neocortex’s computational principles. Future AI systems can incorporate these principles to overcome the stated limitations of current systems. Finally, the article concludes that AI researchers and neuroscientists should work together on specified topics to achieve biologically constrained AI with human-like capabilities.
- 20 July 2021
- Published by Springer Nature in SN Applied Sciences
- Vol. 3 (8) , 1-14
Open AccessA thousand brains: toward biologically constrained AI
This paper reviews the state of artificial intelligence (AI) and the quest to create general AI with human-like cognitive capabilities. Although existing AI methods have produced powerful applications that outperform humans in specific bounded domains, these techniques have fundamental limitations that hinder the creation of general intelligent systems. In parallel, over the last few decades, an explosion of experimental techniques in neuroscience has significantly increased our understanding of the human brain. This review argues that improvements in current AI using mathematical or logical techniques are unlikely to lead to general AI. Instead, the AI community should incorporate neuroscience discoveries about the neocortex, the human brain’s center of intelligence. The article explains the limitations of current AI techniques. It then focuses on the biologically constrainedThousand Brains Theorydescribing the neocortex’s computational principles. Future AI systems can incorporate these principles to overcome the stated limitations of current systems. Finally, the article concludes that AI researchers and neuroscientists should work together on specified topics to achieve biologically constrained AI with human-like capabilities.
- Open Access
- 7 January 2021
- Published by Springer Nature in Computing
- Vol. 104 (1) , 73-93
Tutorial on systems with antifragility to downtime
An antifragile system of software and stakeholders, including designers, developers, and operators, learn from incidents how to avoid outages and maintain high uptime. This tutorial article reviews how to design and operate such socio-technical systems with antifragility to downtime. It documents the importance of four design principles and two operational principles by exploring the polar opposite anti-principles and the interplay between the principles and the anti-principles. The design principles mandate a software design of separate and isolatable processes with sufficient diversity and redundancy. The processes should communicate asynchronously over an external network. The operational principles imply that the software development teams should repeatedly inject artificial failures into the production system to understand its behavior and detect and mitigate vulnerabilities as the system and its environment change.
- 7 January 2021
- Published by Springer Nature in Computing
- Vol. 104 (1) , 73-93
Open AccessTutorial on systems with antifragility to downtime
An antifragile system of software and stakeholders, including designers, developers, and operators, learn from incidents how to avoid outages and maintain high uptime. This tutorial article reviews how to design and operate such socio-technical systems with antifragility to downtime. It documents the importance of four design principles and two operational principles by exploring the polar opposite anti-principles and the interplay between the principles and the anti-principles. The design principles mandate a software design of separate and isolatable processes with sufficient diversity and redundancy. The processes should communicate asynchronously over an external network. The operational principles imply that the software development teams should repeatedly inject artificial failures into the production system to understand its behavior and detect and mitigate vulnerabilities as the system and its environment change.
- 29 July 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computer
- Vol. 52 (8) , 72-75
Biologically Driven Artificial Intelligence
Although deep learning has led to substantial advances in image and speech recognition, language translation, and game playing, we argue that the Al community should focus less on methods using nonbiological ANNs and more on the computational principles of the brain to create human-level or general Al with a performance close to humans at most cognitive tasks of interest.
- 29 July 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computer
- Vol. 52 (8) , 72-75
Biologically Driven Artificial Intelligence
Although deep learning has led to substantial advances in image and speech recognition, language translation, and game playing, we argue that the Al community should focus less on methods using nonbiological ANNs and more on the computational principles of the brain to create human-level or general Al with a performance close to humans at most cognitive tasks of interest.
- 17 June 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Software
- Vol. 36 (4) , 81-87
Dominating Software Systems: How to Overcome Online Information Asymmetry
Today, computer science graduates start companies to create market-dominating, Internet-scale software solutions for hundreds of millions of users. These entrepreneurs want to achieve the success of digital corporations such as Google, Facebook, or Amazon. There is an information asymmetry between current Internet-scale software systems and their users. Although these dominating systems know much about their users, the users know very little about the systems. This asymmetry makes it hard for people to understand the disadvantages of using dominating software systems. In particular, there is an unknown or hidden risk that the systems will cause massive job losses as they move into more and more markets. Because existing Internetscale systems negatively impact society, start-ups as well as established companies should consider the societal impact when they design and grow Internet-scale solutions.
- 17 June 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Software
- Vol. 36 (4) , 81-87
Dominating Software Systems: How to Overcome Online Information Asymmetry
Today, computer science graduates start companies to create market-dominating, Internet-scale software solutions for hundreds of millions of users. These entrepreneurs want to achieve the success of digital corporations such as Google, Facebook, or Amazon. There is an information asymmetry between current Internet-scale software systems and their users. Although these dominating systems know much about their users, the users know very little about the systems. This asymmetry makes it hard for people to understand the disadvantages of using dominating software systems. In particular, there is an unknown or hidden risk that the systems will cause massive job losses as they move into more and more markets. Because existing Internetscale systems negatively impact society, start-ups as well as established companies should consider the societal impact when they design and grow Internet-scale solutions.
- 20 March 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computer
- Vol. 52 (2) , 23-31
Software Systems With Antifragility to Downtime
The authors discuss how to develop and operate large distributed software systems that embrace failures, learn from them, and adapt to maintain a very high system uptime.
- 20 March 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computer
- Vol. 52 (2) , 23-31
Software Systems With Antifragility to Downtime
The authors discuss how to develop and operate large distributed software systems that embrace failures, learn from them, and adapt to maintain a very high system uptime.
- 5 August 2008
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Security & Privacy
- Vol. 6 (4) , 14-20
Open Wireless Networks on University Campuses
Open wireless networks raise privacy issues and entail increased risk of malicious attacks and illegal downloading activities. Such networks are nonetheless attractive-particularly to universities-because they enhance usability and thus expand access to nonsensitive system resources. At universities, such access brings numerous benefits to students, faculty, and the surrounding community alike. Here, the authors describe the challenges of removing individual user authentication requirements at the perimeter of a university network in which mobile device users access system resources over wireless links to the wired infrastructure. The authors discuss how to mitigate the security and privacy risks entailed in an open network of this sort, and also describe how IT departments can vary the network's degree of openness.
- 5 August 2008
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Security & Privacy
- Vol. 6 (4) , 14-20
Open Wireless Networks on University Campuses
Open wireless networks raise privacy issues and entail increased risk of malicious attacks and illegal downloading activities. Such networks are nonetheless attractive-particularly to universities-because they enhance usability and thus expand access to nonsensitive system resources. At universities, such access brings numerous benefits to students, faculty, and the surrounding community alike. Here, the authors describe the challenges of removing individual user authentication requirements at the perimeter of a university network in which mobile device users access system resources over wireless links to the wired infrastructure. The authors discuss how to mitigate the security and privacy risks entailed in an open network of this sort, and also describe how IT departments can vary the network's degree of openness.
- 1 June 1992
- Published by Wiley in Basic & Clinical Pharmacology & Toxicology
- Vol. 70 (6) , 397-401
Interactions between Substance P and Norepinephrine in the Regulation of Nociception in Mouse Spinal Cord
This study examined interactions between effects of the undecapeptide substance P and norepinephrine and the α‐adrenoceptor agonist clonidine in the mouse spinal cord. All compounds were injected into the lumbar subarachnoid space, and effects on the tail‐flick reflex and the behavioural response to intrathecal substance P were evaluated. The tail‐flick response latencies were markedly increased 5–20 min. after intrathecal application of norepinephrine (0.0125–0.1 μg) or clonidine (0.0125–0.1 μg). The actions of both intrathecal norepinephrine (0.025 μg) and intrathecal clonidine (0.025 μg) were significantly attenuated when substance P (5 μg) was given intrathecally 55 and 45 min. before the agonists. There was a significant relationship between the tail‐flick response latencies and the tail skin temperature. However, the tail‐flick results were not due to changes in the skin temperature. Intrathecally applied substance P (10 ng) produced a response consisting of biting of the caudal part of the body and a few hindlimb scratches. The number of bites was significantly reduced 5 min. after injection of norepinephrine (0.1 μg) or clonidine (0.05–0.1 μg), but the number of scratches was unchanged. The data show that increased stimulation of spinal α‐adrenoceptors inhibits a spinal nociceptive reflex as well as the action of substance P in the spinal cord. Substance P modulates the action of α‐adrenoceptor agonists on the tail‐flick reflex, which may tentatively be explained by downregulation of α‐adrenoceptors by substance P.
- 1 June 1992
- Published by Wiley in Basic & Clinical Pharmacology & Toxicology
- Vol. 70 (6) , 397-401
Interactions between Substance P and Norepinephrine in the Regulation of Nociception in Mouse Spinal Cord
This study examined interactions between effects of the undecapeptide substance P and norepinephrine and the α‐adrenoceptor agonist clonidine in the mouse spinal cord. All compounds were injected into the lumbar subarachnoid space, and effects on the tail‐flick reflex and the behavioural response to intrathecal substance P were evaluated. The tail‐flick response latencies were markedly increased 5–20 min. after intrathecal application of norepinephrine (0.0125–0.1 μg) or clonidine (0.0125–0.1 μg). The actions of both intrathecal norepinephrine (0.025 μg) and intrathecal clonidine (0.025 μg) were significantly attenuated when substance P (5 μg) was given intrathecally 55 and 45 min. before the agonists. There was a significant relationship between the tail‐flick response latencies and the tail skin temperature. However, the tail‐flick results were not due to changes in the skin temperature. Intrathecally applied substance P (10 ng) produced a response consisting of biting of the caudal part of the body and a few hindlimb scratches. The number of bites was significantly reduced 5 min. after injection of norepinephrine (0.1 μg) or clonidine (0.05–0.1 μg), but the number of scratches was unchanged. The data show that increased stimulation of spinal α‐adrenoceptors inhibits a spinal nociceptive reflex as well as the action of substance P in the spinal cord. Substance P modulates the action of α‐adrenoceptor agonists on the tail‐flick reflex, which may tentatively be explained by downregulation of α‐adrenoceptors by substance P.