Selected AI Conference Presentations by Dr. Craig A. Kaplan

Dr. Craig A. Kaplan gives a keynote presentation on Safe SuperIntelligence at the AIM (AI and Machine Learning) conference in Seattle, WA, on June 9, 2025. Early in the talk, Kaplan conducts an informal survey of the AI researcher audience, asking for their subjective probability that AI will cause human extinction, known as p(doom). About half the audience estimates p(doom) at 50% or higher, with the vast majority placing it at 20% or higher. Kaplan notes that if researchers could reduce p(doom) by even 0.1%, the expected value of that effort would be 8.2 million lives saved. He then outlines the current paradigm for building advanced AI, including AGI and SuperIntelligence, and highlights six major safety concerns. To illustrate these issues, he uses stories, thought experiments, and simple simulations. Kaplan concludes with a high-level overview of an alternative approach he believes is both safer and faster for developing SuperIntelligence. While the keynote didn’t allow time for technical details, he provided references to ten white papers containing hundreds of pages of implementation guidance for those interested in deeper exploration. (https://www.superintelligence.com/si-research-whitepapers) Key takeaways: 1. SuperIntelligence and other advanced AI should be designed to be safe, rather than relying on testing to uncover issues after models are built. 2. AI will learn from humans - our guidance, values, and behavior - so we must be intentional teachers and role models. While concern about the risks of advanced AI is growing, Kaplan offers a scientific and engineering-based path forward. His keynote addresses the central problem in AI safety today: most researchers simply don’t know how to design safe advanced AI. He presents a solution grounded in better design. Contact: iqinfo@iqco.com https://www.linkedin.com/in/craigakaplan https://x.com/Aligned_SI #superintelligence #agi #aiagents #collectiveintelligence #artificialintelligence #pdoom

AGI 24 (Seattle, August 13-16, 2024). Dr. Craig Kaplan presented A Collective Intelligence Approach to Safe AGI in the session "Consciousness, Creativity, and Trust." If Artificial General Intelligence proves to be a “winner-take-all” scenario where the first company or country to develop AGI dominates, then the first AGI must also be the safest. The safest, fastest, path to AGI may be to harness the collective intelligence of multiple AI and human agents in an AGI network. This approach has roots in seminal ideas from four of the scientists who founded the field of AI: Allen Newell, Marvin Minsky, Claude Shannon, and Herbert Simon. Extrapolating key insights and combining them with the work of modern researchers, illuminates a fast and safe path to AGI. The seminal ideas discussed are 1) Society of Mind (Minsky), 2) Information Theory (Shannon), 3) Problem Solving Theory (Newell & Simon), and 4) Bounded Rationality (Simon). 1. Society of Mind describes a collective intelligence approach that can be used with AI and human agents to create an AGI network. 2. Information Theory helps address the critical issue of how an AGI system will increase its intelligence over time. 3. Problem Solving theory provides a universal framework that AI and human agents can use to communicate efficiently, effectively, and safely. 4. Bounded Rationality helps us better understand not only the capabilities of SuperIntelligent AGI but also how humans can remain relevant where the intelligence of AGI vastly exceeds that of its human creators. Each key idea can be combined with recent work in the fields of Artificial Intelligence, Machine Learning, and Large Language Models to accelerate the development of a working, safe, AGI system. @SingularityNET #superintelligence #ai #llm Websites: https://www.superintelligence.com/, https://www.iqstudios.net/, (video) and https://www.iqco.com (AGI consulting)

A presentation by Dr. Kaplan at the International Conference on Artificial Intelligence and Machine Learning (AIM 2024) in San Francisco on June 11, 2024. After providing some context on the current debate between AI "doomers" and AI "boomers," Kaplan suggests that the only practical path forward is to design AGI to be safe. Fortunately, the fastest and most profitable path to AGI could be the safest if appropriately designed. Kaplan then describes a novel collective intelligence approach to AI safety. His background working with the scientists who pioneered the field of AI, as well as his decades-long industry experience designing and implementing collective intelligence systems, causes him to be optimistic about the prospect of building human-aligned AGI quickly and safely. #artificialintelligence #artificialgeneralintelligence #ai Websites: https://www.superintelligence.com/, https://www.iqstudios.net/, (video) and https://www.iqco.com (AGI consulting)

Catch the highlights in this unofficial video of the ABET 2024 Symposium’s Industry Advisory Council’s AI Panel Session. Moderated by Dr. Naguib Attia, Distinguished Engineer and Vice President (Retired), IBM Global University Programs, panelists were Matt Conwell from Invenergy LLC, Dr. Craig Kaplan of iQ Company, and Andrew Lloyd from AECOM. Topics addressed included: AI's implications for various industries The specific needs industries have for AI technologies Expectations from educational bodies like ABET in advancing AI competencies Audience questions sparked further exploration into AI safety, the phenomenon of AI 'hallucinations', and the pursuit of sustainable AI practices. More about ABET, the panel, and the symposium: https://symposium.abet.org/ #ai #artificialintelligence #IndustryAI #AIandIndustry #EducationalStandards #AIChallenges #PanelHighlights #AIinEducation #FutureOfWork #EngineeringEducation #AIEthics #paneldiscussion Websites: https://www.superintelligence.com/, https://www.iqstudios.net/, (video) and https://www.iqco.com (AGI consulting)

In this unofficial recording from the ABET Symposium 2024, the panel “Ethical Dimensions of AI: Navigating Inclusive Transformations in Education and the Future of Work” was led by Sam Cummings of DataPlant. Panelists included Dr. Craig Kaplan, CEO of iQ Company; Dr. Karen Panetta, Tufts University's Dean of Graduate Education for the School of Engineering; and Kevin Kuck, a mechanical engineering senior from George Mason University. Discussions centered on the unique opportunities and ethical considerations presented by AI. The panelists underscored the significance of upholding data integrity, ensuring safe AI practices, and creating appropriate standards and metrics. A dynamic Q&A with attendees enriched the conversation. Discover more about the panel, the symposium, and ABET by visiting ABET.org. #ABETSymposium #ABET2024 #ethicalai #aiineducation #futureofwork #engineeringeducation #aiethics#paneldiscussion #ai #aitools #generativeai #techinnovations Websites: https://www.superintelligence.com/, https://www.iqstudios.net/, (video) and https://www.iqco.com (AGI consulting)

How to Create AGI and Not Die is a conference presentation by Dr. Craig A. Kaplan, delivered at the IFoRE / Sigma Xi Conference on November 10, 2023. Kaplan provides an overview of AGI and explains the safety challenges, current approaches to safety, and his view of the future. Kaplan explains the necessity of being first with the safest AGI and outlines a path forward. He also describes what all of us, AI experts or not, can do to improve AI ethics. #ai #agi #aisafety #alignmentproblem #artificialintelligence #aiethics #aivalues @smartereveryday Websites: https://www.superintelligence.com/, https://www.iqstudios.net/, (video) and https://www.iqco.com (AGI consulting)