by Malcolm Murray

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. They are: Risk Identification, Risk Assessment, Risk Evaluation, Risk Treatment and Risk Governance. These go by different names depending on the framework, and are sometimes also grouped differently. For example, in the International AI Safety Report, we call them Identification, Analysis/Evaluation, Mitigation and Governance. In a recently published paper that I worked on, Open Problems in Frontier AI Risk Management, they are instead Risk Planning, Risk Identification, Risk Analysis, Risk Evaluation and Risk Mitigation.

The exact wording and grouping does not matter, what matters is the activity. I have found that a good way to explain what the key activities are can be done using the concept of the map and the territory. Basically, first, we want to understand which territory we are in – this is risk identification. Then we create a map of that territory – that is risk assessment. Third, we determine alternative paths to proceed through the territory. This can be compared to risk evaluation. We then need to choose a path and pursue it. This is risk treatment. And finally, the risk governance equivalent aspect can be expressed as stopping once in a while to get a second pair of eyes on the compass, to determine that we’re still going in the right direction.

However, in what has become the norm in current AI risk management, only a limited subset of these are actually being done. What typically happens is an assessment of capabilities, and then, if the capabilities reach a specific threshold, a corresponding implementation of predetermined safeguards.

This means there are several missing areas where risk management is not happening. Read more »

by Laurie Scheck

1.

Always when I think of him, I picture him alone in a room, bedridden for months at a time. How from late childhood on, he suffered from osteomyelitis, an incurable infection of the bone that set flame to raw nerve fibers, induced recurring fevers, skin lesions, intense, crippling pain. And how in that enforced solitude he wrote about the ways the self is not isolate but comes into being through contact with others. How this is its beauty and its freedom. “I become myself only when revealing myself for another, through another, and with the help of another.”

In his great work on Dostoevsky, he wrote, “Every thought…senses itself to be from the very beginning a rejoinder in an unfinalized dialogue. Such thought is not impelled toward a well-rounded, finalized whole. It lives a tense life on the borders of someone else’s thought, someone else’s consciousness.”

Bedridden and alone in his room, far from the clamor of everyday life, he wrote of how language moves through one body to another, one mind to another, and in that movement changes and is changed.

2.

These are some of the rooms he lived in:

An apartment in Kustani, a town in northwestern Kazakhstan where he was sentenced to internal exile for “subversive activity.” Its climate was so harsh and its winds so strong there were cables to grab onto while walking. Here he reported once a week to the security police and was forbidden to teach school; the charges against him involved corrupting the thoughts of the young. He found work as a bookkeeper. His only known publication from this period was an article for the periodical, Soviet Trade, where he discussed economic indicators, including such minutiae as how many pairs of galoshes were ordered by the local collectives. This sentence lasted from 1930-1934. Read more »

Einstein & Etta James

I’m fourteen when Einstein
is in Princeton

He’s old & heading out of breath
as I am fresh and new at breathing
in some

But there the landscapes of the space
between our ears divides

Al used levers of his mind for lifting
mass in time, while mine
took to bass guitar of R & B
which thumped in head
and loins in adolescent
instants

Einstein’s gone, but his poster’s on our wall
—if you were stepping down our stairs you’d
glimpse him in overcoat and hat
strolling through our stairwell, wry smile,
moustache, life-sized poster, white and black,
shedding light on light (you couldn’t miss him)

Same thing, mid-fifties, bluesy
Etta James, to me, was winsome

Strong and sexy
beautiful and sure,

So I weighed the heft of
his and hers and took them both
to heart and mind, and in terms
of back-and-forth between the two,
there certainly has been some

I took to both, and
then some

by Jim Culleny
12/15/18

Enjoying the content on 3QD? Help keep us going by donating now.

by Eric Feigenbaum

On Christmas Eve 1965 – roughly five-and-a-half months after the nascent Singapore’s independence – “The Mexicans” arrived. Their job: to help build and train a Singaporean military. At the time, the British remained nominally committed to protecting Singapore as part of a planned military phase-out, but Singapore needed its own defensive capabilities.

Today it’s easy to look at Singapore and wonder what enemies it could possibly have. But in 1965, Singapore was surrounded by hostile powers – most specifically Indonesia and Malaysia, which for slightly different reasons were resentful, if not opposed to Singapore becoming its own country.

Malaysia was a scorned lover who at the last minute prior to independence from Britain tossed Singapore out of the Malaysian Federation largely for refusing to subscribe to the “Bumiputera First” movement – a set of legal principles that among Malaysia’s multi-ethnic population, gives additional rights and privileges to the indigenous Malays, or “Sons of The Soil”.

Indonesia’s founding leader and dictator, Sukarno – who styled himself with a mononym like Socrates, Charlemagne, Cher and Madonna – believed the Malaysian Federation (later Malaysia and Singapore) was a “neocolonialist” plot by the British to encircle and disempower Indonesia. He particularly resented formerly British Sabah and Sarawak on Borneo going to Malaysia.

As a result, roughly 1.9 million people, 75 percent of whom were Chinese ethnic living on a 500 square kilometer island, relying on water from their hostile northern neighbor, having insufficient agriculture to self-sufficiently feed themselves and being surrounded by enemies found themselves vulnerable. Read more »

by Charles Siegel

Last week, on Presidents’ Day, President Trump lost in court. This is hardly news. The administration has already lost hundreds of cases in just a little over a year. And this isn’t the first time it has lost on a day when courts aren’t even open.

The ruling came in a case officially entitled City of Philadelphia v. Burgum. Of the many, many cases the administration has lost, this is not remotely the most important. The day before I submitted this column, for example, the Supreme Court ruled against Trump in the tariffs case. (Trump, naturally, immediately called the justices who ruled against him — including three of the six conservatives on the court, two of whom he had appointed and all of whom had given him wide and unprecedented immunity —  “fools and lapdogs” and an “embarrassment to their families.”)  And just the day before that, a federal judge in West Virginia granted a habeas corpus petition filed by an ICE detainee. He began his opinion as follows:

Antiseptic judicial rhetoric cannot do justice to what is happening. Across the interior of the United States, agents of the federal government—masked, anonymous, armed with military weapons, operating from unmarked vehicles, acting without warrants of any kind—are seizing persons for civil immigration violations and imprisoning them without any semblance of due process. The systematic character of this practice and its deliberate elimination of every structural feature that distinguishes constitutional authority from raw force place it beyond the reach of ordinary legal description. It is an assault on the constitutional order. It is what the Fourth Amendment was written to prevent. It is what the Due Process Clause of the Fifth Amendment forbids.

In our constitutional republic, governmental force derives its authority from the Constitution. But that authority is not unlimited. The Government’s power is legitimate only because it is derived from the People and exercised through law by identifiable public officers answerable to the public and to the courts. The structure of the Constitution guarantees visibility. Both the officer and the force he employs are traceable to authority delegated by the People and subject to the limits imposed by law. When the Government uses force against the public, the citizen can recognize the officer as a lawful representative. The public can evaluate the act. The judiciary can later review it. Every stop, arrest, detention, and use of force can be tested against the Constitution’s protections. Not so here. Read more »

by Rafaël Newman

If she hadn’t died in July 1987, at the age of 81, my maternal grandmother would be celebrating her 120th birthday this week. One-hundred-and-twenty is a proverbial age among Ashkenazi Jews, who recite the Yiddish formula Biz hundert un tsvantsik, or “[You should live] to 120,” on the anniversary of a birth. That’s a full 20% markup over their Goyish compatriots in the Old Country, who grant honorees a mere Sto Lat!— “100 years”—in the traditional Polish song of birthday congratulations.

I never once said Biz hundert un tsvantsik to my grandmother, however, on any of her birthdays, although not for want of wishing her well. This was because, despite her birth in the canton of Aargau, the traditional Swiss “Pale of Settlement” for Jewish residents until the late 19th century, and her own eventual close family ties with Jews, my Grossmutti was not herself Jewish. But neither was she in fact a Swiss citizen, despite the place of her birth. She was German.

Eva Bertha Strassner was born on February 23, 1906, in Baden, Switzerland, to Johann Simon Emil Strassner and Maria Elisabeth Sander, immigrants from Löbelstein, which was, at the time of their births and until 1918, in the Duchy of Saxe-Coburg-Gotha, and is today a part of Bavaria. Eva Bertha’s father had come with his burgeoning family in search of employment as a handyman, which he found with a variety of enterprises in the industrialized regions of Switzerland at the turn of the 20th century. Over the next decades the family moved, in keeping with Johann Strassner’s fortunes and economic exigencies, from Baden in the canton of Aargau to nearby Wettingen, then to Töss, at the time a village in the canton of Zurich, now a district of Winterthur, the canton’s second city, and from there, in the 1920s, to Schwerzenbach, on the edge of Zurich’s Oberland, within shouting distance of Zurich itself, the cantonal capital and today Switzerland’s largest city. Read more »

A new pedestrian bridge across the river Eisack in Franzensfeste, South Tyrol.

Enjoying the content on 3QD? Help keep us going by donating now.

by David Winner

Jessie Buckley, an incredible actress, breathes life into the role of Shakespeare’s wife, Agnes, in Hamnet. Like others in the film, her face is always a little dirty (Chloé Zhao, the director, doesn’t fall for the trap of giving characters in dirty times perfectly clean skin), but I was frustrated by her character, which struck me as an odd fusion of romanticized gender essentialism and wild irrationality. Though I (a middle-aged cis male living in Brooklyn, New York, in the present day) could not be further from the Elizabethan woods where the essence of Agnes’s character originated, I still found myself resisting the film’s portrayal of her. Spoiler-alert, this writing reveals the ending.

Born (rumor has it) from a wood witch’s womb, Agnes mixes odd herbs together to create tinctures to heal wounds and reduce fevers. She (and this is a nice touch) greets young William’s desire to “handfast” with her by jumping past the wedding to the wedding night and having sex with him animalistically in nearly plain view. And she gives birth by herself in the woods, real wild-woman style. Her character seems ever wise and in touch with nature, her feelings and instincts spiritually sacrosanct and nearly unassailable—a vision of femininity pretty impossible, I would imagine, for any actual woman to live up to.

Her prophecies don’t always seem correct, however. She predicts a future for her son working on plays with William, though he ends up dying, still a child, when he takes the pestilence about to kill his sister onto himself—a brother bravely and spiritually sacrificing himself to save his sister.

Before her son’s death, Agnes encourages Shakespeare (who is sweet and kind of happy-go-lucky) to go to London to follow his passion. She declines his invitation to move there with the family, uncomfortable, I would imagine, in the large, alienating city. Each time William visits Stratford and leaves to return to London, there is tremendous sadness, which reflects—or so I imagine—not just their upcoming separation but the fragility of those times, when the smallest illness, in a time of plague, was likely to carry someone off. Read more »

by Philip Graham

I’m still amazed that Kipling Knox was my student during the mid-to late 1980s, the earliest days of my teaching at the University of Illinois. A more-than-promising creative writing undergraduate student, he then went on to live his professional life in the tech and editing worlds before embarking on a literary career that so far has produced two excellent books—a story collection and a novel—set in the Illinois landscape where he grew up. A landscape also filled with actual ghosts, a popular condition of so much imaginative writing set in the Midwest. My own latest novel has a dance card filled with ghosts, so it was inevitable that Kip and I would eventually sit down and compare notes. Another point of comparison has been our individual decisions to explore the possibilities of independent publishing and the unexpected avenues such a decision can lead authors in search of autonomy. It seems we had a lot of say, and so have divided our chat into two digestible parts.

Philip Graham: In the past three years you’ve published two beautifully-written works of fiction. In 2022, Under the Moon in Illinois, and at the end of 2025, How to Love in a World Like This. Both books are set—largely—in a Midwestern town that you’ve conjured up and call Middling, Illinois. And so they seem to be a part of a larger—and still growing?—world-building enterprise of fiction.

Kipling Knox: Thanks, Philip. Yes, that’s true—both books share a world with common characters. But that wasn’t my original intent. Between publishing these two, I started two other novels, with different settings. I put them both aside because I found myself drawn back to Middling. The story “Downriver,” in particular, ended so ambiguously that I was curious to know what would happen to its characters, Morgan and Arthur, and how their mystery would play out. It’s a difficult trade-off—sticking with one fictional world versus exploring others. When you write a book, you are deliberately not writing others, and there can be a sense of loss in that. But it’s very gratifying to explore a world you’ve built more deeply. I think of how a drop of ocean water contains millions of microorganisms, each with their own story, in a sense. So the world of Middling County (and also, in my second book, Chicago) has infinite potential for stories!

As I consider my next project, I do feel very committed to the American Midwest. Midwestern people and culture fascinate me. I think this is partly because Midwesterners are so often unburdened by a sense of superiority. We’re always trying to prove ourselves, politely. This is a gross generalization, of course, but I think it’s grounded in truth. These qualities make Midwestern characters nuanced and earnest and, hopefully, a little comical. It also makes for rich social satire—I think of midwestern-born authors like Mark Twain, Kurt Vonnegut, and George Saunders. Read more »

by Brooks Riley

Enjoying the content on 3QD? Help keep us going by donating now.

Sughra Raza. Finding Color. Boston, January, 2026.

Digital photograph.

Enjoying the content on 3QD? Help keep us going by donating now.

by William Benzon

The digital computer emerged out of devices that were built during World War II for cracking codes and calculating artillery tables. Almost as soon as the computer came into existence people were thinking about using it for chess and language. Alan Turing developed Turochamp, an algorithm for playing chess in 1948. Though that algorithm was never implemented, chess became a central topic of AI research, so much so that John McCarthy, who coined the phrase, “artificial intelligence,” would eventually write an article entitled “Chess as the Drosophila of AI.” In 1949 Warren Weaver, then with the Rockefeller Foundation, wrote a memo proposing machine translation. The first public demonstration of machine translation took place in 1954 at the IBM head office in New York and involved 60 Russian sentences.

These two lines of research differed on a philosophical level. AI was a full-on assault on human intelligence. Chess was regarded as the apogee of human intelligence. If we could program a computer to play chess at a championship level, we could program a computer to do anything a human can do. That’s what researchers believed. That’s why chess was so important as a domain of research.

The goal of machine translation was more modest: the reliable translation of documents in one natural language (e.g. Russian) into another natural language (e.g. English). The research programs were correspondingly different and wouldn’t merge/collide until the 1970s with the Speech Understanding Project sponsored by ARPA (the Advanced Research Projects Agency of the Defense Department, now simply DARPA, Defense Advanced Research Projects Agency).

However, I do not intend to recount the history of research in these two areas. As I said, my aim is more philosophical. I’m interested in the radically different nature that these two cases present for AI research. To that end I want to begin by discussing the radically different geometric footprints presented by chess and natural language. With that ground under our feet we can go on to more abstract matters.

Read more »