Toppling Epstein’s intellectuals network

While there have been no other high-profile exits from the MIT Media Lab after Ethan Zuckerman and J. Nathan Matias submitted their resignations, the lab’s students had been demanding its director Joi Ito to resign over his ties with Epstein. While it is ridiculous that Ito pled ignorance in his August 15 note where he admitted he had received money from Epstein for the lab as well as as investments in his personal projects, tweets by Xeni Jardan and others only made his ignorance more implausible.

Peter Aldhous and his colleagues at BuzzFeed subsequently used tax filings to track down many of his elusive grantees in one frighteningly long list that includes biologists Martin Nowak and Robert Trivers as well as the publisher of Nautilus magazine.

According to a new set of updates that hit the news over the weekend, Ito had been letting on less than he knew, and he knew that Epstein was a convicted sexual offender who had preyed upon young, vulnerable women for his sexual pleasure as well as that of a bevy of celebrities (including Marvin Minsky, the cofounder of the Media Lab). The following articles – led by Ronan Farrow of The New Yorker, who apparently published the first article based on whistleblowers at MIT who had known of Ito’s and others’s (non-ignorant) ties with Epstein but whose notes the New York Times had turned down, possibly because Ito is on the Times‘s board of directors – have all the details:

  1. Jeffrey Epstein’s Donations Create a Schism at M.I.T.’s Revered Media Lab (NYT)
  2. How an Élite University Research Center Concealed Its Relationship with Jeffrey Epstein (NYer)
  3. Director of M.I.T.’s Media Lab Resigns After Taking Money From Jeffrey Epstein (NYT)
  4. The Epstein scandal at MIT shows the moral bankruptcy of techno-elites (The Guardian)

There is also this…

… and this (the whole thread is excellent):

https://twitter.com/AnandWrites/status/1169949412371705856

Farrow goes into great detail in his story but the most revealing paragraph to me was this:

… the lab was aware of Epstein’s history—in 2008, Epstein pleaded guilty to state charges of solicitation of prostitution and procurement of minors for prostitution—and of his disqualified status as a donor. They also show that Ito and other lab employees took numerous steps to keep Epstein’s name from being associated with the donations he made or solicited. On Ito’s calendar, which typically listed the full names of participants in meetings, Epstein was identified only by his initials. Epstein’s direct contributions to the lab were recorded as anonymous. In September, 2014, Ito wrote to Epstein soliciting a cash infusion to fund a certain researcher, asking, “Could you re-up/top-off with another $100K so we can extend his contract another year?” Epstein replied, “yes.” Forwarding the response to a member of his staff, Ito wrote, “Make sure this gets accounted for as anonymous.” Peter Cohen, the M.I.T. Media Lab’s Director of Development and Strategy at the time, reiterated, “Jeffrey money, needs to be anonymous. Thanks.”

While it was already ridiculous at the time of Ito’s first indication that he accepted Epstein’s money without knowing of Epstein’s crimes, it is absolutely certain now that Ito spent many, many years knowing what Epstein had done and expressed regret for his actions only when the heat became unbearable.

What’s more, MIT and the Media Lab are guilty of the same thing, descending to the moral cesspit occupied by universities around the country , and the world, that harboured exploitative professors who harassed their students, and purchased their employers’ silence with scientific expertise – whatever that stands for – and federal grants. This outcome also supports the view that without the right sociological safeguards, the naked scientific enterprise is hugely vulnerably to being instrumentalised to achieve extra-scientific goals. And Cesar Hidalgo, a former associate professor at the Media Lab and then its first and sole Hispanic member, said in a thread recounting his experiences that Ito had done just this, in his own way.

(Aside: Whenever a scientist is informed that he or she is a suspect in a crime in the TV show Elementary, their first response is often along the lines of: “But I’m a scientist!” I tend to burst out laughing at this point. It is fascinating how many people believe scientists are to be perceived as incapable of committing crimes by virtue of being scientists, as if they are not people too and – more importantly – as if they are people enslaved to the diktats of the natural universe and whose directions they follow in an unbiased and unemotional manner.)

Earlier, on August 22, Evgeny Morozov published an intriguing article in the New Republic, in which he shared an email he received from John Brockman in 2013 that showed Brockman knew about Jeffrey Epstein’s criminal activities as he continued to associate with him, and even tried to recruit intellectuals to interacting with him.

Brockman runs Brockman Inc., a literary agency that represents the who’s who of intellectual authors and writers, including Morozov himself, and now helmed by his son. More importantly, Brockman is the man behind the Edge Foundation, which runs Edge.org, an internet salon of sorts where he invites some of the world’s more renowned scientists and philosophers to discuss their ideas. Edge also hosts an annual event for the world’s billionaires, called ‘The Billionaires’ Dinner’.

Morozov’s contention was that Brockman has been awfully silent about his ties with Epstein, even though it has come to light that many of the intellectuals in Epstein’s orbit were launched there by Brockman, as well as that Epstein donated $638,000 (Rs 4.5 crore) to the Edge Foundation between 2001 and 2015. Morozov apparently fired Brockman Inc. as his literary agency until the man could clarify what his relationship with Epstein was, and emailed the notice to Brockman’s son, who currently runs the company, and shared that email on Twitter on August 26:

Morozov also encouraged other Brockman clients to speak up, and sever ties if need be with him, his agency and/or his foundation. While only a few people answered his call, it is to the whistleblowers’, Farrow’s and the Miami Herald‘s credit that being or having been associated with Epstein is finally acknowledged as a problem that isn’t subject to individual moral codes but is being recognised as an incontestable evil. I hope it is only a matter of time before more scientists recognise this, and subsequently that greater participation from their own ranks in the efforts to understand S&T’s role in society is the best way to keep such Epsteinian affairs from recurring in future.

Elementary P v. NP

“What’re you currently binge-watching?” is a great conversation-starter, I think. One well-suited for this age and carrying enough insights into the person you’re trying to strike up a conversation with.

I’m currently binge-watching Elementary, an adaptation of Sherlock Holmes whose mysteries may not be as well-roundedly gripping as they are in Sherlock but whose narrative spares me the melodrama of Benedict Cumberbatch and whose reality is neither as ambitious nor paranoid. I also like the idea of a male-female pair in the lead that doesn’t have sex from the get-go.

But in its simplicity, Elementary sometimes get ahead of itself. The most common symptom of this is that, in a few episodes, there is some information that is revealed in the second-half, to which the viewer has had no previous access and so to whom the resolution veritably arises from the blue. As a once-avid quizzer, this is a buzzkill because it offers no opportunity for the viewer to attempt to solve the puzzle before Holmes does (which isn’t impossible). It also makes a lesser ‘deductionist’ of Holmes because, in the viewer’s eyes, it robs him of the ability to piece together a solution to a puzzle using pieces the viewer knows exist.

But in some other episodes, another kind of problem arises. In them, Elementary ventures into fantasy. Often it has the good sense to explain some conundrums away with the suffix “… if we had a really well-equipped lab and billions of dollars”. At other times, it assumes the viewers ignorant. And at even other times, it clubs the two.

An example of the last variety is season 2, episode 2: ‘Solve for X’. In this episode, two mathematicians are killed while they’re both working on a solution to the P versus NP problem. The problem, first formally presented by a computer scientist named Stephen Cook in 1971, asks whether P equals NP. P stands for the set of all problems that can be solved in a reasonable amount of time (relative to their complexity). NP stands for the set of all problems that cannot be solved in a reasonable amount of time but that can be verified in a reasonable amount of time.

(John Pavlus had the perfect metaphor for NP back in 2010: “Imagine a jigsaw puzzle: finding the right arrangement of pieces is difficult, but you can tell when the puzzle is finished correctly just by looking at it”.)

Though a solution to the problem has evaded the most skilled mathematicians for many decades, most of them intuitively believe that P ≠ NP. Solving the problem fetches a $1 million cash prize from the Clay Mathematics Institute (CMI), New York – setting up the motive in ‘Solve for X’. However, the episode is also careful to note the fact that most of modern cryptography is problem-solving and that a solution to the P v. NP problem would be worth hundreds of millions to digital security companies.

This is a line that was bandied about just last month when a group of physicists announced that they were a step closer to resolving the Riemann hypothesis, another mathematical problem that carries a $1 million reward from the CMI. Resolving the hypothesis has some beautiful implications for the incidence of prime numbers on the number line and for prime factorisation, both integral components of the RSA encryption algorithm. However, it says little about how a problem itself could be solved – a notion that is also applicable to the pop culture surrounding the P v. NP problem.

In ‘Solve for X’, a third mathematician solves the problem and then uses it to hack emails and into CCTV footage.

Proving that P equals NP does not mean programmers immediately have insights into where the solutions for a given problem may lay. They only know that, technically, the problem has a solution in polynomial time. For example, consider the RSA algorithm itself. It is effective because it relies on prime factorisation, the process of finding two large prime numbers whose product is being used as a key to cracking the security the algorithm affords. A single-core 2.2-GHz CPU is supposed to take “almost 2,000 years” to factorise a number with 768 bits. Numbers with 2,048 bits or higher are thought to be unfactorisable in a single human lifetime. At the same time, given two prime numbers, their product can be quickly computed and a proposed solution rapidly verified.

But proving P equals NP doesn’t make any of this easier. Knowing that there is a solution out there is not the same as knowing what it could be, particularly in situations where scientists are concerned with practical applications.

R.J. Lipton, an expert on the problem, probably has the most meaningful take on this: that succeeding in solving the problem is about making a psychological impact. If P ≠ NP, then your conception of the world isn’t shaken much: some problems that you thought were hard are going to remain hard. But if P = NP, then it means there are a whole bunch of problems that you thought were hard but are actually easy, and that you – and thousands of mathematicians over the years – have missed out on simpler answers. However, it wouldn’t have anything to say about where this ease arises from.

… unless there is either something encoded in the solution of the P v. NP problem itself or it’s the case that a unified solution emerges to crack all NP problems. I felt that ‘Solve for X’ takes such encoding for granted, that it significantly overestimates the confidence boost accorded by knowing that a problem is solvable or that it significantly underestimates the gap between P = NP and being able to hack into everything. In other words, if P equalled NP, I would still be aghast if someone hacked into Google’s servers.

Any which way, it goes a step too far, evident when you see Tanya Barrett and her programmer friend go from solving the problem to hacking into CCTV footage within a day. It makes a fool of a viewer by saying that this is what the P v. NP problem immediately helps accomplish – and it also makes savants of Barrett and the programmer in order to be able to make such rapid progress but whose savantism doesn’t manifest in any other way.

SPOILERS END

I also think Elementary may have missed a great opportunity to explore a wider, wilder world in which P = NP.

As with my criticism of Spectral, my taking on of Elementary happened because it was an interesting intersection of science and popular culture and that doesn’t happen often. And while Elementary may not have perfectly depicted the consequences of solving a famous mathematical problem – just the way Spectral took some liberties with depicting an esoteric physical phenomenon – I’m glad that it ventured in that direction at all.

Finally, if you’re interested in reading more about the P v. NP problem, I highly recommend Scott Aaronson’s and Lipton’s blogs. Both can get quite technical but they also have some posts that discuss the problem very well, by which I mean not just simple language but also with insights that might allow you to glimpse the underlying principles. Here are two posts – here and here – to get started with.

Featured image credit: Pexels/pixabay.