Topic
Gary Marcus
24 stories related to this topic, newest first.
benzinga.comai44 days ago
Anthropic Revenue Projection to $2 Trillion by 2030 Illustrates Growth Fallacy
AI researcher Gary Marcus said that forecasts of Anthropic reaching $2 trillion in annual revenue by 2030 exemplify the trillion pound baby fallacy. He compared such projections to assuming a baby that doubles in weight during its first four months will continue doubling at the s…
GA
IBrazal (WMF) / Wikimedia (CC BY-SA 4.0)technology71 days ago
Gary Marcus Reports Unreliability in X Platform's Search Function
Gary Marcus stated that the search feature on X is unreliable. He suggested that an undergraduate could develop a superior system for classifying trends. Marcus noted that X operates as part of a company involved in rocket and AI sectors.
GA
Substrate placeholder — needs reviewtechnology71 days ago
Angel Investor Calls for Violence Against AI Critic Gary Marcus
An angel investor has called for violence against Gary Marcus, a known AI skeptic. Marcus has stated that he opposes violence and supports boycotts instead. The incident involves efforts to misrepresent Marcus's positions on AI and related issues.
GA
Substrate placeholder — needs reviewai72 days ago
Gary Marcus States Opinion on Limitations of Large Language Models
Gary Marcus, a researcher in artificial intelligence, stated that large language models (LLMs) are of limited use. He described the associated hype as ill-founded and potentially dangerous. Marcus presented this view as his professional opinion based on his experience as a practi…
GA
Substrate placeholder — needs reviewtechnology72 days ago
Gary Marcus Questions Sam Altman's Stated Moral Obligations in AI Development
Gary Marcus, an AI researcher, has publicly questioned Sam Altman's commitment to his stated moral obligations in AI, citing concerns over surveillance and intellectual property practices. Altman, CEO of OpenAI, previously described working toward prosperity for everyone, empower…
GA
Substrate placeholder — needs reviewai72 days ago
Gary Marcus Agrees with Dean Ball on Need for AI Regulation
Gary Marcus stated agreement with much of Dean Ball's views on AI regulation. He emphasized that belief in imminent AGI is not required to support prompt regulatory development. Such regulation would aim to foster innovation while addressing potential risks.
GA
Substrate placeholder — needs reviewai72 days ago
Gary Marcus Comments on Garry Kasparov's Chess Skills Versus Large Language Models
Gary Marcus stated that Garry Kasparov retired from competitive chess over 20 years ago, with most of his tournament games publicly available. Marcus suggested that Kasparov could still outperform any large language model in chess without specialized tools or training. The remark…
GA
Substrate placeholder — needs reviewtechnology72 days ago
Gary Marcus Applies Garry Kasparov's Statement on Data Privacy to AI Leaders
Gary Marcus referenced a statement by Garry Kasparov regarding assurances on personal data privacy. Kasparov expressed doubt about a claim that personal conversations would not be sold to the government. Marcus extended this skepticism to leaders in the AI industry.
GA
Substrate placeholder — needs reviewai72 days ago
Gary Marcus Comments on LLM Narrative Consistency
Gary Marcus, an AI researcher, stated that he would have less to say if the LLM community adhered to a specific narrative. This remark appears in a social media post. The comment highlights ongoing discussions in AI development.
GA
Substrate placeholder — needs reviewai72 days ago
Comparison of Regulatory Frameworks for Candles and Artificial Intelligence
Gary Marcus has stated that candles face more regulatory oversight than artificial intelligence systems. He noted that candle manufacturers do not appear to seek exemption from liability for product failures. This observation highlights differences in industry approaches to regul…
GA
Substrate placeholder — needs reviewtechnology72 days ago
Gary Marcus Critiques Extreme Capitalism in Relation to AI Risks and Eliezer Yudkowsky's Warnings
Gary Marcus has expressed concerns about extreme unfettered capitalism in the context of artificial intelligence development. He links this economic approach to potential existential risks that align with warnings from Eliezer Yudkowsky. The statement highlights ongoing debates o…
GA
Substrate placeholder — needs reviewtechnology73 days ago
Gary Marcus Questions Mythos Model in Washington Cybersecurity Discussions
Gary Marcus stated that policymakers in Washington have been influenced regarding the Mythos cybersecurity model. He argued that Mythos does not represent an effective approach for addressing cybersecurity concerns. Marcus referenced his newsletter for detailed reasons supporting…
GA
Substrate placeholder — needs reviewtechnology73 days ago
Gary Marcus Comments on Neuroscience Knowledge in Public Discourse
Gary Marcus, a cognitive scientist, stated that certain behaviors do not indicate understanding of current neuroscience realities. The comment appears in the context of ongoing discussions about AI and brain science. No specific individual or event was named in the statement.
GA
rte.ieai73 days ago
Gary Marcus Reports Decline in Generative AI Popularity
Gary Marcus stated that the popularity of generative AI has reached a plateau. He noted challenges for OpenAI in meeting financial targets. Marcus also referenced OpenAI's acquisition of a media company in this context.
GA
Substrate placeholder — needs reviewai73 days ago
Google DeepMind CEO Comments on AI Priorities in Hypothetical Scenario
Demis Hassabis, CEO of Google DeepMind, stated in a discussion that if decisions were his, efforts would have focused on curing cancer before developing consumer AI tools like ChatGPT. Gary Marcus referenced this comment to highlight concerns over AI development paths led by Open…
GA
Substrate placeholder — needs reviewai73 days ago
Gary Marcus Comments on Anthropic's Mythos Announcement and Its Limitations
Gary Marcus stated that Anthropic's announcement of Mythos on the previous day did not represent a significant advancement. He noted that the demonstration involved turning off sandboxing, which limited its relevance to real-world applications. Marcus also pointed out that simila…
GA
espn.co.uktechnology73 days ago
Gary Marcus Assesses Mythos Cybersecurity Claims as Largely Unsubstantiated
Gary Marcus stated that details surrounding the Mythos cybersecurity incident are mostly inaccurate. He described the event as a warning for improved cybersecurity measures. Marcus advised against taking specific details at face value.
GA
Substrate placeholder — needs reviewtechnology74 days ago
Gary Marcus Assesses Mythos Cybersecurity Incident as Largely Unsubstantiated
Gary Marcus stated that the Mythos cybersecurity incident was mostly a myth. He described it as a warning sign for improving cybersecurity measures. Marcus advised against taking the reported details seriously.
GA
Substrate placeholder — needs reviewtechnology74 days ago
Gary Marcus Comments on AI Developments and References Ramez Naam's Thread
Gary Marcus stated that recent AI gains do not represent exponential progress or recursive self-improvement. He described 'Mythos' as not an off-trend exponential gain. Marcus recommended a thread by Ramez Naam on the topic.
GA
Substrate placeholder — needs reviewtechnology74 days ago
Gary Marcus Comments on Untagged Critics and Argument Flaws
Gary Marcus posted on social media about individuals who avoid tagging him in discussions due to perceived weaknesses in their arguments. He invited followers to identify flaws and misrepresentations in a referenced argument. The post frames this as an intellectual exercise.
GA
Substrate placeholder — needs reviewtechnology74 days ago
Gary Marcus Offers Balanced Perspective on Mythos AI Security Incident
Gary Marcus, an AI researcher, published a post discussing the Mythos AI system, suggesting it may not be as problematic as some reports indicate. He referenced insights from cybersecurity expert Heidy Khlaaf, who has audited safety-critical systems. The post aims to provide cont…
GA
Substrate placeholder — needs reviewtechnology74 days ago
Gary Marcus References 2025 Politico Article on Cyberattack Vulnerabilities Updated to 2026
Gary Marcus stated that a Politico article predicting heightened U.S. vulnerability to cyberattacks in 2025 accurately describes the current situation in 2026. He suggested changing the year in the article from 2025 to 2026 to reflect present conditions. Marcus described the U.S.…
GA
allthingsdistributed.comtechnology75 days ago
Gary Marcus Comments on AI Field's Progress in Temporal Reasoning Since 2019
Gary Marcus, author of the 2019 book Rebooting AI, stated that the AI field has spent seven years on an architecture unable to address temporal reasoning. He described this situation as both hilarious and unsurprising based on discussions in the book. Temporal reasoning remains a…
GA
Substrate placeholder — needs reviewai75 days ago
AI Tools Used to Adapt Social Media Content for Higher Engagement
Gary Marcus reported that social media algorithms prioritize content based on engagement levels. He noted that large language models (LLMs) enable users to rewrite posts and news stories into versions that attract more interaction. Marcus observed this practice occurring frequent…
GA