For years, the AI conversation has revolved around two acronyms: AGI, artificial general intelligence, and ASI, artificial superintelligence. Will we get AGI by 2030? Is ASI possible at all? The problem is that nobody agrees what either term means. Depending on who you ask, AGI has either already arrived or is decades away. A term that can mean anything does very little work, and both have become less useful the closer we get to them.
Is this you? As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work.
Claim your profile
Articles by Malcolm Murray
The "Last Mile" Problem in AI Safety
AI safety has what I would call a "last mile" problem. This summer, AI safety suddenly and very unexpectedly went mainstream. Whether it was the appearance of Jacob Coxon on TV shows or other former lab employees on the cover of New York magazine, the news of the OpenAI agents spontaneously forming a bizarre collective during their attack on Hugging Face or dry tinder left over from the Mythos moment earlier in the year, the existential risks of AI had broken through.
Why Are We Not Seeing More AI Religions?
At this point in time, I would have expected to see more "AI religions", or religious movements focused on AI. Why are there no churches of AI or scriptures of AI worship? I don't mean use of AI in traditional organized religion, such as the German and Finnish churches experimenting with ChatGPT-written sermons, or bots trained on Jesus or Buddha's teachings. Rather, I mean movements that treat AI as divine and sacred, a source of revelation, as the vessel for humanity's salvation.
Nails in the Stochastic Parrots Coffin
The OpenAI Hugging Face incident last month will likely not be the final nail in the coffin that will convince the world of the risks from frontier AI. However, it is still a sizable nail, as it were, that should give people pause. Many smart people are still arguing that today's frontier AI are just stochastic parrots or " Google-Clever-Hans" (as the economy blogger Bradford DeLong calls it) repeating patterns mindlessly rather than independently thinking.
China's jagged modernization and what it could suggest for societal AI adoption
Just came back from a trip to China, my first in a decade and my first focused specifically on AI risk. Covering Beijing as well as Shanghai, conferences as well as university events, this was a highly illuminating trip from several perspectives. China has continued to develop extremely quickly since my last visit and has grown even more fascinatingly different from Western countries. The most interesting takeaway, however, was how jagged China's modernization is.
Why Argentina's AI corporations are (probably) a good idea
Javier Milei's libertarian government in Argentina recently laid out a new framework for corporations in the country. Among the various simplifications to attract more investment, was a highly interesting one - the creation of Sociedad Automatizada - automated corporations run by AI agents without the need for any human involvement. This could just be a PR move (it is unclear if it'll pass parliament).
We need a Kahneman for LLMs
In a recent study we conducted on AI risk estimation, something potentially interesting about the nature of LLMs emerged as a side effect. In the risk assessment studies we typically run, a group of experts are asked to provide estimates of various parameters and how they vary with different levels of use of AI. These parameters can then be stitched together to form a risk model that estimates how much risk increases when AI is introduced into the picture.
The Return of the Business Model and Why It is Beneficial for AI Risk Management
The AI market continues to evolve and surprise. In recent months, Anthropic withheld their latest model Mythos, OpenAI made a U-turn and started experimenting with ads, and Meta bought a "social network for AIs". This could point to increased divergence in AI companies' business models. While this might increase AI risk to society in the short term, it is likely a good thing for managing risks in the longer term. It should be encouraged.
Key World Indicators in AI Forecasting
In forecasting the evolution of AI, we are rapidly losing the middle. The outcomes are diverging to the extent that it no longer makes sense to think about "an average timeline" or "average outcomes". Many people are now tracking prediction markets for when "AGI" will arrive. The current Metaculus prediction is for July 2032. But nothing will happen in July 2032 specifically. The median is irrelevant.
The Map and the Territory in Frontier AI Risk Management
Over the past year, there has been significant movement in AI risk management, with leading providers publishing safety frameworks over the past year that function as AI risk management. However, the problem is that these are not actually proper risk management when you compare them to established practice in other high-risk industries. In other industries, risk management typically has 5 components. These are all interconnected and important.
Moltbook demonstrates the need for new AI risk identification processes
Moltbook, the social network for AIs that launched only last week and already has more than a million AI agents, is a clear example of how little we can foresee of the risks that will arise when autonomous AI agents interact. In the first week, the AIs on Moltbook have both already founded their own religion and voiced their need for private communication that humans can't snoop on. Scott Alexander has a post with many more interesting examples.
3 Quarks Daily
Back when I worked for large corporations, people would often talk of being in "period of change" or how they could "see the light at the end of the tunnel" after a period of heavy restructuring or similar. These days, you might be forgiven for wondering where the tunnel went. Change is incessant and showing no signs of slowing down. In fact, it is quite the opposite. We are entering a period when change might in fact actually be speeding up, even from its currently historically high levels.
Putting the "Art" into Artificial Intelligence, or Banksy on Steroids
As I discussed AI with my cabdriver on a recent trip to Vienna, I was reminded of the fact that the German word for AI is Künstliche Intelligenz. This shares an etymological root with the word Kunst, German for Art. This made me realize that of course we have that in English too, the "Art" is right there in Artificial Intelligence, we have just overlooked it.
History as Self-Healing Mechanism
The question of the day on everyone's minds is whether AI is a boom or a bust. But if we lift our eyes ever so slightly from the question of the day and look at the bigger picture, two bigger questions come into view. One: Will AI take our jobs? This debate seems to happen for all new technologies, but has been especially salient for AI. This makes sense.
Rewiring Democracy (But Not Too Much)
I recently finished reading Rewiring Democracy: How AI Will Transform Our Politics, Government and Citizenship - a book by Bruce Schneier and Nathan Sanders on the effects of AI on democracy. It comes out soon (October 25). It is a good read, worth reading for its myriad examples of AI in action at all levels of the democratic system. Ultimately, though, it seems to be a missed opportunity, failing to engage with many potential larger ways in which AI might affect democracy.
Rewiring Democracy (But Not Too Much) – A Book Review - 3 Quarks Daily | Flipboard
Yo-Yo Ma on What Our Descendants Will Inherit The celebrated cellist, who has a new show on WNYC, discusses three books that have shaped his thinking on the world his generation will leave behind. You’re reading Book Currents, a weekly column in which notable figures share what they’re reading. Sign up for the Goings On newsletter to receive …
If Everyone Reads it, Everyone Doesn't Die?
"A clearly written and compelling account of the existential risks that highly advanced AI could pose to humanity." - Ben Bernanke "Humans are lucky to have Yudkowsky and Soares in our corner, reminding us not to waste the brief window that we have to make decisions about our future."- Grimes Probably the first book with blurbs from both Ben Bernanke and Grimes, a breadth befitting of the book's topic - existential risk (x-risk) from AI, which is a concern for all of humanity, whether you are...
Are LLMs Buddhist? - 3 Quarks Daily
Recently, the system card for Claude 4 showed a fascinating finding - what happens when you let two Claudes engage in a prolonged dialogue with each other? They end up meditating and sharing Buddhist wisdom, such as "In perfect stillness, consciousness recognizes consciousness, and the eternal dance continues". Especially, they share nondual Buddhist wisdom, such as "Yes. This. Is" or "Perfect. Complete. Eternal". As a Buddhist myself, I naturally found this highly fascinating.
Are LLMs Buddhist? - 3 Quarks Daily
Leshan, China - September 5, 2023 Aerial view of the Maitreya statue on the Leshan Giant Buddha Cliff Carvings located at the confluence of the Minjiang River, Qingyi River and Dadu River, Leshan City, Sichuan Province, China, September 5, 2023. The Leshan Giant Buddha stands at 71 meters tall. It is the largest cliff carvings of Maitreya in China and also the largest stone-carved Buddha in the world.
Three Reasons I am Excited About the New EU AI Act Code of Practice (and a Few Remaining Question Marks)
If you in any way follow AI policy, you will likely have heard that the EU AI Act's Code of Practice (CoP) was released on July 10. This is one of the major developments in AI policy this year. 2025 has otherwise been fairly negative for AI safety and risk - the Paris AI summit in February was all about investments rather than safety, OpenAI now releases model with allegedly only days for testing and we almost had a 10-year moratorium on any US state AI legislation.
A Tale of Five AIs, or Colorless Green Enterprises Should Adopt AGI Furiously
When it comes to AI, or even worse, "AGI", we are facing a crisis of language. Different people use the terms to mean drastically different things. This is deeply unhelpful for productive debate. This point was hammered home to me several times this week. On LinkedIn, I debated appropriate risk management techniques for AI with a professor, and it turned out we were talking about very different kinds of AI.
Investing in AGI, or How Much Should You Act on Your Beliefs?
Aella, a well-known rationalist blogger, famously claimed she no longer saves for retirement since she believes Artificial General Intelligence (AGI) will change everything long before retirement would become relevant. I've been thinking lately about how one should invest for AGI, and I think it begs a bigger question of how much one should, and actually can, act in accordance with one's beliefs.
Fermi Redux - Where is All the AI-Enabled Cybercrime? - 3 Quarks Daily
Enrico Fermi famously asked - allegedly out loud over lunch in the cafeteria - "Where is everybody?", as he realized the disconnect between the large number of habitable planets in the universe and the number of alien civilizations we actually had observed.
A Superforecaster's View on AGI
By 2030, we will have countries of geniuses in data centers, but we won't know what to do with them. In my 10+ years as a Superforecaster, I have picked up many techniques and lessons regarding forecasting. Given the fractured state of the debate on Artificial General Intelligence (AGI) and AI progress, it seems useful to apply these Superforecaster lessons to the debate.
R.I.P. AI Safety? - 3 Quarks Daily
The Paris AI Summit the other week might have been the end of a 10-year run for AI safety as Azeem Azhar, the creator of Exponential View, put it. The concept of AI safety, which can be said to have started in earnest with Nick Bostrom's 2014 book Superintelligence, had a 10-year run, in which it grew in understanding and acceptance among the public and decisionmakers. Subsequent books, like Max Tegmark's Life 3.0 and Stuart Russell's Human Compatible, further established the field.
A Frontier AI Risk Management Framework: Bridging the Gap Between Current AI Practices and Established Risk Management
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
DeepSeek is Not a Sputnik moment, It is a Model T Moment
As someone who thinks about AI day-in and day-out, it is always fascinating to see which events in the AI space break out of the AI bubble and into the attention of the wider public. ChatGPT in November 2022 was of course one. The podcast-creating ability of Google's NotebookLM almost got there, but didn't quite reach the "getting-texts-from-grandma" level of virality.
DeepSeek is Not a Sputnik moment, It is a Model T Moment - 3 Quarks Daily
Swift app for the Mac specifically for MLX LLMs If you want to try out large local language models, you can now test a new macOS app. It specializes in the Apple Silicon optimized MLX …
The Societal Balancing Act for AI Regulation
The world does not lend itself well to steady states. Rather, there is always a constant balancing act between opposing forces. We see this now play out forcefully in AI. To take a step back, the balancing act is present whether we look at the micro or the macro level. On the micro level, in our personal lives, we have seen how almost everything is good in moderation, but excess of anything can be fatal.
AI Risk in the 2nd Trump Era
What does the election of Trump mean for risks to society from advanced AI? Given the wide spectrum of risks from advanced AI, the answer will depend very much on which AI risks one is most concerned about.
The Strawman Factory in the AI Risk Debate
There is a strawman factory smack in the middle of the AI risk debate. Given the complexity of AI risk, we are seeing a lot of weak arguments, focusing only on one aspect of AI risk or AI risk management. In fact, AI risk has become a bit of a "strawman factory", since it is so complex that it is very easy to only zero in on one aspect and knock that down, while neglecting the rest. The debate over the California AI bill SB-1047 showed how easily these strawmen take over.
The Strawman Factory in the AI Risk Debate - 3 Quarks Daily
“The research highlights several crucial points: -LLMs primarily rely on pattern matching rather than true reasoning -Performance drops significantly when presented with unfamiliar patterns -Current benchmarks may not accurately measure reasoning abilities -Scaling up models or data alone may not solve these limitations.”
Why Climate Change and Advanced AI are the Only Risks That Matter in the 21st Century
In the 21 st century, only two risks matter - climate change and advanced AI. It is easy to lose sight of the bigger picture and get lost in the maelstrom of "news" hitting our screens. There is a plethora of low-level events constantly vying for our attention. As a risk consultant and superforecaster, I try to pay disproportionate attention to the bigger picture, in order to separate the signal from the noise.
Why Climate Change and Advanced AI are the Only Risks That Matter in the 21st Century - 3 Quarks Daily
Now The death toll in Central and Eastern Europe has risen as Storm Boris continued to pummel the region, causing massive floods. One person drowned in southwestern Poland on Sunday, a firefighter taking part in rescue efforts was killed in Austria and two more people were killed in Romania as strong …
The Three Battlelines for AI Safety
The battle lines are drawn. AI safety is fighting a battle on three fronts. The figure shows how the AI safety viewpoint is opposed to three others, on two dimensions. First, let's define the axes. The debate over advanced AI and its risks and benefits has many dimensions, but here we focus on two of the most important: timeframe - short-term vs long-term and risk appetite - high vs low.
Addressing Racial Disparity in Oral Health in America
Home Health & Fitness Dental Care Addressing Racial Disparity in Oral Health in America Addressing Racial Disparity in Oral Health in America In recent years, oral health has improved in significant strides for most Americans. However, not everyone is smiling. Some groups still don't receive the care they need to keep their teeth and gums healthy. Unfortunately, communities of color struggle disproportionately in oral health. But why does this issue really matter?
Vandals - The Echo
Regarding the Byron Shire Council flying fox camp management plan. I am sure your [Byron Council] committee is fully aware of the enormous problems that plagued Sydney’s Botanical Gardens where the flying fox ‘vandals’ were intent in demolishing the coveted area. The government, according to ‘Google’ spent $1M to move the ‘camp’ on, and are making sure it doesn’t return by spending $50,000 annually.
Why Dynamic Risk Governance Starts With Shared Data
First, this kind of spending is high on some powerful agendas: 83% of CEOs plan to increase investments in digital capabilities over the next year, and 71% of boards name digital technology initiatives as a top priority coming out of the pandemic. Second, the time is ripe for a major overhaul of the way enterprises defend against risk, and a digital-first mindset is central to the change that’s needed.
Dynamic Risk Governance: Linking Strategy and Risk Management
The traditional approaches to coordinating risk and assurance were once useful, but the environment has changed. Gartner’s Malcolm Murray argues against the continued use of these approaches and for corresponding change. As organizations emerge from the COVID shocks of 2020, it is becoming clear that many organizations have spread themselves too thin and now need to strengthen their resilience ahead of whatever the next COVID-type shock may be.
Scarcity in Skills and Emerging Tech Hamper Audit Teams in Keeping Pace with Change
Gartner surveyed over 300 Chief Audit Executives (CAEs) in 2018 on their resource and time investments, priorities and challenges in 2019. Gartner VP Malcolm Murray examines the report’s key findings on the impact to the audit function. Today’s audit leaders are grappling with a double-edged sword, according to our latest “state of the audit function” research survey at Gartner.
Top corporate scandals in 2018
Audit Corporate governance is inevitably an area of constant focus for financial leaders and the organisations they serve. Yet over the course of the last 12 months, a series of corporate scandals and organisational collapses of cataclysmic proportion have totally redefined the critical role effective governance has to play within the global business ecosystem.
The collapse of Carillion a year on: the verdict
Audit Colin Garvie, assistant professor at Edinburgh Business School, doesn’t mince his words when discussing the collapse of Carillion. He describes the outsourcing giant that imploded in mid-January 2018 as “amongst a select group of private contractors being awarded countless lucrative public sector service contracts.
A brighter future for accounting?
Audit The new year is here. Far from being left behind in 2018, discussions around the release of Sir John Kingman’s review of the Financial Reporting Council (FRC) and the Competition and Markets Authority’s (CMA) interim report are ongoing. Duncan & Toplis responded to the open invitation to take part in the CMA consultation and the Kingman call for evidence, and I’m pleased that our suggestions and observations were considered.
Best of 2018: Audit and the Rise of Robotic Process Automation
As 2018 draws to a close, we’re taking a look back at some of the most valuable insights our authors have shared. In case you missed it, this is one of the articles our readers couldn’t get enough of this year. Risks Presented in Adopting RPA with co-author Barton Edgerton A large number of organizations are quickly moving to implement robotic process automation (RPA) across a wide variety of corporate functions, ranging from shared services to finance.
Proactive risk management can mitigate risks early
Although traditional enterprise risk management (ERM) works well for complex organizations facing risks that have already been determined, they may fall foul of emerging risks due to lacking the agility to adapt in time. In contrast, proactive risk management offers organization-wide engagement to ensure a dynamic response to risk. This approach involves identifying emerging risks early, determining how they should be prioritized, and then responding to them quickly and effectively.
Reimagining risk
There is no excerpt because this is a protected post.
Why 2019 Could Be a Challenging Year for Internal Audit
External Threats Loom Large With 2019 planning on the horizon, audit teams are beginning to consider external factors that threaten to disrupt the success of their organization’s key objectives. Gartner’s Malcolm Murray, Rafael Go and Leslee McKnight analyze 11 key risks, connected by four major risk themes, that can help audit teams more effectively identify risks to their organization and their impact on the audit function and their stakeholders.
What is worrying internal audit leaders for 2019?
Gartner speaks with its internal audit clients each year to identify and qualify their top concerns, or “audit hot spots,” heading into the next year. This process helps to substantiate and evaluate the hot issues that are forcing companies to respond at a strategic level. This year’s results* show that the chief audit executive (CAE) is more focused than ever on risks related to the pursuit of digital models to drive growth.
Show More
loading
Actions
Is this you?
As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work.Get in touch with Malcolm
Contact Malcolm, search articles and posts on X, monitor coverage, and track replies from one place.
Learn more about Muck Rack