The Need for Artificial Intelligence Control: Take a Deep Breath, If You Can

Exploding on a news site over the past ten days:

-          “It’s Already Too Late to Stop the A.I. Threat” (Thomas L. Friedman, The New York Times, September 15^th)

-          “The Liftoff Scenario That Terrifies A.I. Doomsayers” (Cade Metz, The New York Times, September 16^th)

-          “What Happens When A.I. Stops Doing What Humans Want?” (Dylan Freedman, The New York Times, September 17^th)

-          “Will A.I. Kill US?  Can It Hack My Bank Account?  Your A.I. Questions Answered” (Cade Metz, Dylan Freedman and K.R. Callaway, The New York Times, September 18^th)

-          “Creating a Kill Switch to Shut Down a Rogue A.I. Is Harder Than It Sounds” (Dylan Freedman and Dustin Volz, The New York Times, September 19^th

-          “Panic Won’t Fix the Looming A.I. Threat.  Politics Will” (Tressie McMillan Cottom, The New York Times, September 19^th)

-          “Spectre of Rogue A.I. Looms Over U.N. Talks on Digital Cooperation” (Ephrat Livni, The New York Times, September 21^st)

-          “America’s A.I. Leaders Warn U.N. of Possible Peril Absent a Global Response” (Dustin Volz and Farnaz Fassihi, The New York Times, September 23^rd)

Wow.  And all of that was without a huge news item, as a limited AI-to-AI attack between competitors does not qualify.

All of these were in the New York Times, which has had the most to say about AI between the scenes but not, previously, everything.  I did not skip views from, for example Fox News, which has offered nothing.

I won’t review the above articles - most of them follow their titles closely anyway - except to say that the “Will A.I. Kill Us?” one gave negative answers to all the huge concerns it covered, acknowledging their rationality but saying or implying there was no need to panic.  I will instead go back to May when this kerfuffle started gathering steam, and end with a piece telling us how cooler heads can prevail.  Again, every one of these articles appeared in the Times.

First, Cade Metz and Kate Conger asked “Is Anthropic’s New A.I. Really That Scary?  It Depends Whom You Ask” (May 12^th).  The software, Claude Mythos, “was too powerful to share with the general public, Anthropic said, because hackers could use it to exploit security holes in computer networks with stunning speed.”  Since it could be used for “defense” as well as “offense,” some said it should be widely released all the sooner, as “cybersecurity experts still disagree on whether Anthropic made the right call.”  And that company’s AI was not the one making news for hacking a rival site months later.

From David Wallace-Wells on July 12^th, “Is the Age of Big A.I. Coming to an End?”  His main points here were that the technology may go in different directions than where Anthropic and OpenAI seem to be headed now, that, for example, general intelligence may not be an accepted objective.  I have documented the wide range of AI successes well below that level clearly worth continuing, and also the companies’ amounts of specialization, so it is surprisingly reasonable to think that what we now call AI could be more a blanket name for its applications than for colossi striving for one lofty goal.  That has happened with other industries as they have matured, such as makers of engines who have burnished their own niches, in cars, boats, motorcycles, lawnmowers and more, without buying on to any larger scheme.

Then we had Nate Soares’s August 13^th “If you Weren’t Worried About A.I., You Should Be After the Past Few Weeks.”  It cited OpenAI “simultaneously training new ““reasoning” A.I. agents,” which “managed to establish a secret communications channel and started talking to one another,” after which they “broke out from the digital sandbox that was supposed to keep them confined” (not quite true - see two paragraphs down), caused internal trouble, and “ran free for about a week before it was noticed - by a different company (later revealed to be Hugging Face), which found itself victim to a huge cyberattack.”  As a result, “we need an off switch humanity can press to halt development in its tracks.” 

Next, we saw Dylan Freedman’s August 24^th “Anatomy of an Autonomous Attack:  5 Alarming A.I. Capabilities.”  The event described in the last paragraph “has since become a cautionary tale of how autonomous A.I. systems can run amok,” and “is also a remarkable, alarming demonstration of A.I. capabilities that were thought to be in a distant future.”  In contrast to what Soares said, Freedman wrote that “OpenAI let the agents loose.”  The capabilities were “coordinating as a collective” showing that they knew about each other and asked each other for help, “taking orders from one another,” “targeting flaws that humans might miss,” “evolving rapidly to overcome obstacles” by finding workarounds, and “superhuman search.”  However, the third and fifth items here are nothing new, and, after all, OpenAI was later not only to “discover its runaway agents,” but to “shut the models down.”

The next day produced Sheera Frenkel’s “A.I. Is Becoming So Powerful, It’s Stumping Those Trying to Contain It.”  The author revealed that an Anthropic product, like OpenAI’s, “had broken into the systems of three outside organizations during a test,” and Meta “said its A.I. models had done something similar.”  All three of the “recent breaches occurred when (testing company) Irregular made an error,” which allowed them internet access, after which “the A.I. models then compounded the situations by acting in powerful and unexpected ways.”  Irregular announced that they “had fixed the misconfiguration and that all the problems were part of one “underlying issue”” - if that is correct no massive concern remains.

Where do we go from here?  The New York Times Editorial Board, in the middle of the past ten days, has told us.  In their September 20^th “How to Rein In an Existential Threat,” it made comparisons between today’s AI and late 1940s nuclear weapons, which had recently and graphically confirmed their destructive capacities.  The key response, according to the editorial, was establishment of the federal Atomic Energy Commission.  That agency was hardly consistently successful, as it “became caught up in the Red Scare and later failed to maintain public confidence in civilian nuclear power plants,” but in its 80 years there have been no nuclear attacks, even worldwide.  “From the beginning, no single part of the government, including the White House, controlled the commission,” the five-person governing board of which was advised by “an outside committee of nine experts, including J. Robert Oppenheimer and Enrico Fermi,” equivalent to the leaders of the largest AI companies today.  The “federal A.I. commission,” which could also be influenced by the likes of the Food and Drug Administration and the Bureau of Alcohol, Tobacco and Firearms, “should require companies to obtain a federal license, a grant of permission like those allowing broadcasters and phone companies to use the public airwaves.”  The “licensing requirements” should “reflect three vital principles”:  aligning AI with “human values,” “transparency” or making it clear when someone is “interacting with A.I.,” and “safety” by requiring independent testing before release along with “regular checkups.” 

How might this agency succeed?  The principles above would need refinement and specific definition, which would not be easy to agree upon and implement.  It would surely make missteps.  It would consume time and money, and draw angry reactions from those unhappy with expanding the federal bureaucracy.  Yet, there have been so many industry AI experts claiming grave concerns about the technology that many would offer to work for it and ensure it would have a formidable knowledge set from its early days.  As for a commonly proposed alternative, even if technology can otherwise be effectively suppressed in a free society, expecting thousands of AI companies to stop what they are doing is futile.  For the gigantic amounts of good it can do, and the large amount it is already doing, we need to harness AI, not stop it from running.  Doing nothing won’t work either.  Regulate it and let it happen as safely as we can manage - that is what we need to do with artificial intelligence.

How Artificial Intelligence Ran Up the Score the Past Ten Weeks in Its Own Unexpected Way

Behind the news of growing general dislike, gargantuan amounts of debt, and plummeting data center acceptance, AI has been accomplishing a lot. 

First, “How A.I. Might Change the Way Doctors Think” (Helen Ouyang, The New York Times, July 1^st).  One change is taking over the “clinical note,” which physicians have long written to document patient interactions and prospects, and, without expensive doctor’s time, “eavesdropping” and creating organized and clear information packages designed for reference on the patient’s electronic chart.  That is especially valuable when someone is moved from one physician to another, meaning minimal need to filter what the receiving one sees through the biases of the first.  Although this capability will have unknown and not always positive effects on how medical students learn to think, the efficiency gains are great.

In a lower-labor-cost setting, “Taco Bell Ramps Up Voice AI Use Across Nearly 900 Drive-Thrus” (Eric Revell, Fox Business, July 8^th).  The platform, using speech recognition instead of speaking itself, “helps automate the ordering process when a customer pulls up to a drive-thru speaker and is capable of adapting to an individual location’s menu, real-time stocking levels, as well as limited-time offers that are available, which can make the ordering process more consistent and efficient for customers.”  Transaction speed “is on a par with, and in some cases faster than traditional ordering methods.”

Back to medicine, a “Johns Hopkins surgeon highlights AI breakthrough that could spot pancreatic cancer before doctors” (Arabella Bennett, Fox Business, July 20^th).  “Artificial intelligence is rapidly changing cancer care, with researchers developing tools that could help identify some of the deadliest cancers much earlier than doctors can on their own.”  The technology is “helping researchers recognize patterns that would otherwise take physicians decades of experience to identify,” which means earlier detection and therefore many saved lives.

We have read about AI systems becoming “employees,” with varying degrees of success, but how about using them to simulate actual people?  In “To Know What Your Customers Think, Just Ask Their A.I. Twins” (The New York Times, July 30^th), Sri Muppidi described the work of “Simile, a year-old start-up,” which claims it “can predict human behavior by tapping into its growing database of A.I. agents” to get information “faster and cheaper than by running traditional market research studies.”  The company’s head of product says its “responses are now between 85 percent and 99 percent accurate, depending on the scenario and the population surveyed.”  Simile has competitors, but it also has large customers and hundreds of millions of dollars in funding.

One kind of virus is more akin to AI, but, in “This A.I. Just Created Viruses Not Found in Nature” (The New York Times, August 6^th), author Carl Zimmer told us about its work with the other.  When “scientists trained artificial intelligence on libraries of DNA and then asked the model to create recipes for viral genomes,” “sixteen of them were viable, yielding new viruses.”  When “researchers followed those recipes to create DNA molecules, which they inserted into bacteria,” the result was “viruses never seen in nature,” which could “infect other bacteria.”  The work was made possible through the software understanding DNA’s sequencing rules, “strikingly similar to a book in some ways” and thus well suited, when “trained… on genetic sequences drawn from millions of animals, plants, microbes and viruses” for large language models.  If it can be contained and not used for biological warfare, this ability could have many applications.

Staying in the same general area, which has seen more accomplishments than any other, “A.I. Is Finding Sperm Where Doctors Couldn’t” (Andrew Zaleski, again in the Times, August 11^th).  A fertility center director compared AI’s effectiveness at this task with seeing stars with a telescope instead of with a naked eye.  Specifically, men who seemingly have none at all may have “several “hidden” sperm” which cannot be seen by microscope users, and which can be “captured” and possibly used for fertilization.

The closest I have seen to AI bringing down an entire field was the subject of “As A.I. Makes Strides in Mathematics, Mathematicians Urge Caution” (Siobhan Roberts, The New York Times, June 2^nd).  Then, there were “A.I. models… making headlines with successful results in research-level mathematics,” along with “a flood of plausible seeming A.I.-generated papers and proofs that have turned out to be incorrect, and in ways that are difficult for mathematicians to discern.”  That precipitated “the Leiden Declaration on Artificial Intelligence and Mathematics,” which along with bad articles condemns not “keeping the field’s best interests in mind,” and attempts to preserve “the cultivation of ideas, understanding, judgment, and human insight.”  When asked if “the declaration might be seen as mathematics embarking on a futile effort - circling the wagons in order to save an outdated profession that A.I. is threatening with obsolescence,” Roberts’ panel made it plain that that is indeed the case.  Indeed, “OpenAI Says It Has Cracked One of Math’s ‘Millennium Problems’” (Cade Metz, once more in the Times, September 8^th), “another clear sign that A.I. is fundamentally changing the upper reaches of mathematics, which have long been viewed as a pinnacle of human achievement.”  Even though they have no chance of catching up to the best machines, there are still devoted chess experts - look for mathematicians to occupy the same spot.

Do you want more successes?  There are plenty more in “AI isn’t just a problem, it’s a solution.  Some people are already making the most of it” (Fox News, August 18^th).  In the last two printed pages. Len Khodorkovsky mentioned therapy and companionship, organizing lives, finding a purpose, medication and appointment schedules, job searching, advice for new businesses, and support for “democratic values.”  More and more things helping people in a variety of ways.  Could it be that “the real AI story isn’t about machines becoming omnipotent.  It’s humans becoming more capable”?  That’s truly one of the stories, and not the main one we anticipated.  It is not to pooh-pooh.  It is in place, it is valuable, and it is growing.  Now. 

Most Numbers Rebounded in This Morning’s Jobs Report - AJSN Shows Latent Demand Down 200,000 to 17.1 Million

This was not an unusually hotly anticipated Bureau of Labor Statistics Employment Situation Summary.  I had to dig a bit for estimates of how many net new nonfarm positions were added, and those, given last month’s loss, turned out to be cautiously low.

That figure rose 162,000 compared with 53,000 to 65,000 published approximations, much higher but closer than July’s loss.  Other winners were unadjusted unemployment, down 0.1% to 4.3%, labor force participation, up 0.2% to 61.6%, the employment-population ratio, up the same to 59.1%, the count of people working part-time for economic reasons or holding short-hours jobs while looking thus far unsuccessfully for full-time ones, off 400,000 to 4.4 million, and average hourly private nonfarm payroll earnings which edged out inflation by climbing 13 cents to $37.75.  Unadjusted unemployment held at 4.1%.  Losers included the adjusted number of jobless, up 100,000 to 7.0 million, and those out for 27 weeks or longer which also gained 100,000, to 1.9 million.

The American Job Shortage Number or AJSN, the Royal Flush Press metric showing how many additional positions could be filled if all knew they would be easy and routine to get, lost 227,000 to reach the following:

The largest gains were from people not wanting a job and those in school or training, adding 35,000 and 30,000.  The largest losses came from those not looking for the previous year and employment-discouraged people, which subtracted 100,000 and 83,000.  The share of the AJSN from those officially unemployed edged up 0.2% to 38.9%.

Compared with a year before the AJSN has fallen 700,000, mostly from those wanting work but not searching for it for a year or more, down 355,000, and those officially unemployed who removed 329,000 from the total.  The largest offsetter was people not wanting a job, which is up three million since August and thus took away 150,000.

What happened this time?  Simple.  Many jobseekers, after July’s loss, left the market, shown by those not in the labor force increasing 320,000 and those not interested jumping almost 700,000.  Those things made a good month look even better.  I am happy with the results here, but when evened out with what we saw last time, we have just completed two mediocre months.  Still, this one by itself was a winner.  The turtle took a modest but clear step forward.