AI, Surveillance, And Curing What Ails Us

A lot of people, I think, get a lot wrong about security and AI. Twenty years ago, when I was briefly involved in the Dartmouth alumni trustee movement, I participated in online discussions among students and alumni that taught me a lot. I discovered that a major complaint among conservative students was that they had to swap electronic ID cards to get in and out of the dorm buildings. They felt this was a restriction on their freedom.

I had a career spent partly on the road, and I thought that electronic key cards in hotels were a major advance in traveler safety. Why would anyone object to something like this? Well, they didn't like the idea that someone could look up where they'd been and when they'd gone in and out.

I tried explaining that a record like that could actually be in their favor -- in an age of unsubstantiated rape accusations, for instance, it could prove you weren't in someone's dorm when they claimed you were. They still didn't like it, and I was beginning to get a feeling it was based on a sense of elite entitlement. I soon enough lost interest in the alumni trustee movement. Alumni trustees weren't going to fix the Ivy problem.

Sundance at Conservative Treehouse is raisimng what I think are related issues with the surveillance state, but his concerns are mostly reasonable.

The concern we have is about the “surveillance state” and the creations of the Silicon Valley tech bros that are building out tools that can easily be weaponized as the contracts with the U.S government are fulfilled. President Trump will not weaponize them, but the existing tools and the DHS track and trace enhancements being built by the tech team are threats to liberty.

. . . What we have researched with Elon Musk carries many of the same background datapoints and nuance. Musk takes popular positions to cover for some very un-MAGA proposals; the H1B issue is just one facet. Overall, a technological system of surveillance and control by a few tech oligarchs is well underway.

Sundance links to an article from last September about Musk's silent partner Larry Ellison, Billionaire Larry Ellison says a vast AI-fueled surveillance system can ensure 'citizens will be on their best behavior'. One thing the article gets wrong is that it assumes these things are in the future; in fact, they're here now.

Walking down a suburban neighborhood street already feels like a Ring doorbell panopticon.

But this is only the start of our surveillance dystopia, according to Larry Ellison, the billionaire cofounder of Oracle. He said AI will usher in a new era of surveillance that he gleefully said will ensure "citizens will be on their best behavior."

There's no future tense about it -- such systems are widely in operation, but why aren't we already on our best behavior? For instance in New Orleans:

The New Orleans Office of Homeland Security boldly protects communities by operating the city’s state-of-the-art, 24/7 Real-Time Crime Center (RTCC). This nerve center proactively monitors locations across the city using advanced technology to provide critical intelligence to first responders in real time.

Opened in 2017 as part of our agency’s $40 million Citywide Public Safety Improvement initiative, the RTCC stands on the cutting edge based on global security best practices for local, state and federal public safety leaders. Analysts access hundreds of CCTV and ShotSpotter feeds while coordinating response logistics down to exact addresses of incidents including crimes and quality of life issues.

Instead of a citizenry cringing in fear of Big Brother, we simply have an increasingly empowered lumpenproletariat, which the surveillance systems struggle to keep up with. A recent episode of the A&E show Homicide Squad New Orleans covers two random and seemingly unrelated street killings that police discovered were committed by a single perp.

Det. Walter Edmond responds to a homicide in Central City, where a young woman named Cayla Kelley has sustained multiple gunshot wounds and is found dead in the street. . . . Edmond is led to a primary suspect named Earl Simmons. Edmond collects surveillance footage that places Simmons at the crime scene with Cayla and issues an arrest warrant. A few days later, a ballistics match shows that the murder weapon that killed Cayla, killed another innocent victim named Eddie Matthews. This piece of evidence breathes new life into Det. Jameson Diesburg’s investigation, who once feared the case would go cold.

The New Orleans RTCC's role in making the footage of two separate shootings in two separate areas quickly available to the detectives was mentioned only in passing. Surveillance footage from traffic cameras, business and home security systems, and doorbell cams has become an important investigative tool. The broad ability of AI technology to consolidate access to such footage is already working to improve law enforcement capability.

The probolem is that right now,things are falling apart. Public spaces and public transit are unsafe. For instance,

Barriers to protect bus drivers have been installed on [Los Angeles] Metro's entire fleet, the transit agency said Tuesday.

The barriers are designed to protect drivers. In April, the Metro board approved an emergency procurement motion to accelerate the installation following an uptick in crime on the Los Angeles transit system, some of which targeted drivers.

. . . The completion of the barrier installation comes at the end of a year that saw several high-profile crimes involving Metro, including a bus hijacking in September in which a passenger was shot and killed. That bus was equipped with a safety barrier designed to prevent a hijacker from approaching the driver and taking control of the bus.

The deadly hijacking was about six months after another Metro bus was commandeered by an armed man in downtown Los Angeles. That bus collided with cars and eventually slammed into the Ritz-Carlton Hotel. The bus driver and a woman in a car struck by the bus were injured.

For now, we can only begin to infer the connection between illegal immigration and increased crime, which primarily affects low-income and minority populations. The public expectation is clearly that Trump will move quickly to reduce the rapidly expanding criminal class, and some form of enhanced data correlation will need to be used to identify aliens with criminal records, locate them, and remove them. I would say that by and large, these methods already exist; the issue is the political will to employ them. It appears this also exists now as well.

Sundance is correct in his assumption Trump won't weaponize these capabilities. In addition, the political process has already demonstrated that weaponizing these capabilities is unpopular. For now, I think the prospect of usimng AI-related capabilities to improve the current public environment outweighs the potential danger.