Don’t forget potential non-existential global catastrophes

AI extinction fears risk overshadowing preventable health threats

Anthony Butler

Apocalyptic fears bring with them a danger that more mundane challenges amenable to policy action will be downgraded

First published in Business Day

15 October 2026

Last month, a researcher called Jacob Coxon resigned from the artificial intelligence firm Anthropic, claiming that, “the people building AI earnestly believe that it could kill us all by the end of the decade.” The company’s Evan Hubinger added that the risk of human extinction within a decade is more than 10%.

The mechanisms bringing existential risk remain speculative, and analysts believe AI giants may favour mild regulation to protect them against insurgent competitors and shield them from massive product-liability exposure when their models enable costly cyberattacks. Nonetheless such warnings come alongside competing alarms about global nuclear war, asteroid collisions, novel pathogens, and synthetic mirror-image organisms bringing untreatable infections.

Such apocalyptic fears bring with them a danger that more mundane challenges amenable to policy action will be downgraded. For this reason, it is good news that the Lancet Commission on 21st Century Global Threats to Health has produced a new report categorizing and analysing dozens of global challenges.

The commission’s analysis builds on a baseline scenario that models future demographic shifts and economic trends to help understand how vulnerabilities will affect different populations.

It assesses “the number of years in good health that are lost due to sickness, disability or premature mortality”, a measure known as lost disability-adjusted life years (Dalys).

The report then categorizes global challenges into distinct tiers, ranging from “extinction potential” risks like nuclear conflict and the malicious use of AI to “major catastrophic” risks such as climate change, antimicrobial resistance, and obesity.

The “Dalys” approach captures well the potential impact of low probability risks as AI-engineered biological weapons, cascading disruptions in critical infrastructure, physical detonation of nuclear weapons, planetary agricultural collapse, and global nuclear famine. After all, if we are all dead, there will be no Dalys.

However, it also gives due weight to “major catastrophic threats” that cost fewer life years but are far more likely to occur. Climate change, for example, will bring direct health impacts from extreme heat, shifting patterns of infectious disease, floods, storms, and climate-fuelled air pollution. The study suggests action to contain global warming to 1.8C rather than allowing it to reach 4.4C could prevent almost two billion years of healthy life from being lost.

While headlines frequently focus on extinction-potential threats like nuclear war and AI annihilation, the study’s quantitative approach reveals that most preventable health loss over the next 70 years is likely to be caused by more everyday metabolic and social conditions.

Tobacco accounts for billions of lost DALYs as the dominant driver of chronic obstructive pulmonary disease, cardiovascular disease, and lung cancer. Smoking rates in low and middle income countries, moreover, continue to grow, and regulation continues to be circumvented.

High Systolic Blood Pressure is meanwhile the leading physiological driver of stroke, ischemic heart disease, and kidney failure. As with smoking, governments have scope to intervene more to limit its costs using well-understood and cheap interventions such as salt intake reduction and antihypertensive medication.

Worst of all is what the commission calls the “Obesity Cluster” of weight, poor diet, and low physical activity, which it describes as the single largest threat to global health. This cluster of pathologies fuels type 2 diabetes, cardiovascular diseases, chronic kidney disease, and cancers. It is driven by ultra-processed foods and sedentary lifestyles that governments can address through policy change.

The commission’s message is that fears about extinction-potential risks should not distract us from well-understood challenges that we are in a position to address using properly funded and “threat-ready” health systems, expanded prevention and preparedness, innovations such as new antibiotics, obesity therapeutics, and climate mitigation, and other tried and tested tools from conventional public policy armouries.

Butler teaches public policy at the University of Cape Town

October 1, 2026

Dangers of AI training

ANTHONY BUTLER | What if AI trains using Mbeki’s speeches?

Argument in a report by former president’s foundation ignores the real issues

First published in Business Day and BusinessLive

1 October 2026

Amid all the brouhaha about the threat that Artificial Intelligence (AI) poses to the future of human life, one even more frightening prospect has hitherto been overlooked. Rather than killing us, AI may instead subject us in perpetuity to the eternal guidance of large language models trained on the collected speeches and writings of former president Thabo Mbeki.

This terrifying spectre has been raised by the Thabo Mbeki Foundation’s release of a new report entitled The United States of America and the Future of Afrikaners in South Africa.

A pithy 123 pages long, and so relatively condensed for a contribution to the Mbeki oeuvre, the paper reinterprets the current disputes between the two countries as a manifestation of the “national question” in both countries.

In Mbeki’s usage, the national question refers to the resolution of colonial conquest’s legacy by freeing the dispossessed indigenous majority and building one united, non-racial nation where race no longer determines wealth or opportunity. The basic parallel the report draws is that the US has faced its own national question exhibited in slavery, segregation and the continuing socioeconomic marginalisation of Black Americans.

From this perspective, the Foundation argues that it is historically inconsistent for Washington to demand that SA abandon racial redress only three decades after 1994 when the US continued to use affirmative-action and other remedial policies long after emancipation.

This is not a very good argument, of course, which is one reason we can speculate that Mbeki played a personal role in the preparation of the report.

There is further evidence of Mbeki’s hand in accounts of transition era negotiations with Constand Viljoen which seem to be presented from his personal perspective, and the text slips into the first person singular when describing Mbeki’s interactions with Inkatha Freedom Party leader Mangosuthu Buthlezi.

To its credit, the foundation assembles empirical material to contest American misconceptions about SA, deploying the analyses of notable Afrikaners Frans Cronje and Herman Giliomee to argue that Afrikaner living standards rose faster between 1994 and 2004 than during any comparable 20-year period since 1945.

This certainly suggests that a genocide or persecution narrative cannot plausibly describe the socioeconomic condition of Afrikaners as a whole (but then we all knew that already).

The report then veers rather alarmingly in an Mbekian direction by equating the building of Afrikaner community institutions with a revival of apartheid-style “separate development”. It also speculates that the current US administration has aligned itself with a “white supremacist faction” in SA, citing as evidence US connections of Solidarity and AfriForum, the refugee programme, and opposition to land reform. There is evidence for each element of this narrative, it is true, but this does not amount to proof of such an alignment.

For now, Mbeki Foundation’s work still clearly draws on Mbeki’s intellect rather than on AI. A sophisticated language model would usefully have placed more weight on Trump’s domestic political coalition, US polarisation over racial preferences, immigration politics, evangelical and conservative networks, SA foreign policy with regard to China and Iran, US commercial interests, and SA’s involvement in the International Court of Justice against the government of Israel – all matters elided by Mbeki’s abstract and characteristically paranoid theoretical generalisations.

It is therefore essential that any future global AI safety plan prohibits the training of language models on archival materials gathered by the Mbeki Foundation. In the centuries ahead, when human beings send robots between the stars in fleets of interstellar rockets, we cannot take the risk that their interactions with alien civilisations will be mediated by Mbeki-trained embodied AI. Such robots will surely antagonise Daleks, Vulcans and Klingons by presenting reports about how to avert threats to the national democratic revolution, and by ruminating at great length on the application of the national question to human-alien relations

Butler teaches public policy at the University of Cape Town

Beware of white ladies

in chemise dresses

and pretty sandals

that show their toes. Beware of these ladies when spring is here.

They have strange habits Of infesting our townships with seeds of:

geraniums pansies poppies carnations. They plant their seeds in our eroded slums cultivating charity in our eroded hearts making our slums look like floral Utopias. Beware!

Beware of seeds and plants.

They take up your oxygen

and they take up your time

and let you wait for blossoms

and let you pray for rain

and you forget about equality

and blooming liberation

Chris van Wyk, It Is Time to Go Home. Johannesburg, Ad Donker, 1979,

President Cyril Ramaphosa meets Chancellor Angela Merkel of the Federal Republic of Germany for bilateral talks ahead of the G20 Compact with Africa and G20 Investment Summit which take place tomorrow 30 October 2018, in Berlin.South Africa, with Germany Co-Chair the Advisory Group on the Compact with Africa that was initiated under Germany’s Presidency of the G20 in 2017.Germany and South Africa enjoy a robust and growing trade relationship.In 2017, Germany remained South Africa’s 3rd largest global trading partner.