There is a very important point to understand about the vendors using ML for threat detection. Usually ML is used to identify known behavior, but with variable parameters. It means that many times we know what bad looks like, but not how exactly it looks like. For example, we know that data exfiltration attempts will usually exploit certain protocols, such as DNS. But data exfiltration via DNS can be done in multiple ways.

Members of Congress, the U.S. military, and prominent technologists have raised the alarm that the U.S. is at risk of losing an Artificial Intelligence (AI) arms race. China already has leveraged strategic investment and planning, access to massive data, and suspect business practices to surpass the U.S. in some aspects of AI implementation. There are worries that this competition could extend to the military sphere with serious consequences for U.S. national security. During the prior Cold War arms race era, U.S. policymakers and the military expressed consternation about a so-called "missile gap" with the USSR that potentially gave the Soviets military superiority. Echoes of gap anxiety continue today.

In an effort to deal with a shortage of radiologists affecting hospitals, several European companies have trained AI to detect signs of breast cancer. Kheiron Medical recently announced that it will use AI algorithms to try to diagnose breast cancer with a trial on historic scans launching this month at a NHS trust in Leeds. The company's technology will also be tested against tens of thousands of historic scans in the East Midlands. Google's own AI company, DeepMind has also recently begun a trial with the NHS and the Dutch company ScreenPoint Medical has developed similar technology. All of these companies are testing their algorithms on tens of thousands of mammograms to determine if they can train them to identify signs of breast cancer with the same competency as a human radiologists.

As the world's oldest and largest open-source intelligence agency, Jane's has spent the last 120 years collecting, classifying and analyzing information. We support national security analysts who are trying to understand the emerging threats to a country's national interest and how best to respond to them. Those analyst teams, our customers, could spend many weeks collecting the information, understanding it, verifying it, and then reporting on it. Or they can work with us to get highly accurate, relevant information from our open and classified sources. Then they can quickly get down to the business of reporting and providing decision-support to their agencies and governments.

If you had to program a self-driving car, which option would you choose if only two were available: hit a pedestrian who suddenly appears in front of the vehicle or veer off into a baby carriage on the sidewalk? It's the kind of ethical conundrum that could shape artificial intelligence in years to come, and one of many the University of Alberta's Geoffrey Rockwell has been pondering lately. Earlier this month, the professor of philosophy and digital humanities joined a national brainstorming forum on the ethics of AI in Montreal, along with industry leaders, federal government officials and other academics, including philosophers. They gathered to grapple with an industry currently worth US$7.4 billion, according to figures circulated at the forum, and expected to reach almost US$16 trillion by 2065--amounting to a seismic shift in how we live and work. The forum followed the signing last June of the Canada-France Statement on Artificial Intelligence, meant to jump-start an international coalition charged with exploring the societal implications of a technology that promises to soon be as ubiquitous as the internet, only with the power to potentially make life-and-death decisions on our behalf.

NOTE: This is an excerpt of Planet Money's newsletter. You can sign up here. Two weeks ago, MIT's David Autor gave the prestigious Richard T. Ely lecture at the annual meeting of American economists in Atlanta. Introduced by the former chair of the Federal Reserve Ben Bernanke as a "first-class thinker" who was doing "path-breaking" work on the central economic issues of automation, globalization, and inequality, Autor strolled up to the microphone with a big smile. His talk was about the past and future of work, and he focused especially on cities.

No one should trust Artificial Intelligence (AI). Trust is a relationship between peers in which the trusting party, while not knowing for certain what the trusted party will do, believes any promises being made. AI is a set of system development techniques that allow machines to compute actions or knowledge from a set of data. Only other software development techniques can be peers with AI, and since these do not "trust", no one actually can trust AI. More importantly, no human should need to trust an AI system, because it is both possible and desirable to engineer AI for accountability. We do not need to trust an AI system, we can know how likely it is to perform the task assigned, and only that task.

Federal regulators have announced plans to allow drone operators to fly their unmanned aerial vehicles over populated areas and at night. A Wing Hummingbird drone from Project Wing arrives and sets down its package at a delivery location in Blacksburg, Va., last year. Federal regulators have announced plans to allow drone operators to fly their unmanned aerial vehicles over populated areas and at night. A Wing Hummingbird drone from Project Wing arrives and sets down its package at a delivery location in Blacksburg, Va., last year. Package delivery by drone is one small step closer to reality today.

Are you jaded by the overuse of Artificial Intelligence (AI) – with vendors instilling either fear or faith? In the cybersecurity domain, we see CISOs investing in Machine Learning (ML), but remaining justifiably skeptical of AI. Machine learning – a building block for AI – lets augmented analytics help security staff decide what to investigate, detect low-and-slow attacks that defenses have missed, and gain enough time to explore the serious problems.ExtraHop Security teams at enterprises still drown in too many warnings. In November, Enterprise Management Associates (EMA) found that 64% of alerts go uninvestigated, and only 23% of respondents investigate all of their most critical alerts. Machine learning – a building block for AI – lets augmented analytics help security staff decide what to investigate, detect low-and-slow attacks that defenses have missed, and gain enough time to explore the serious problems. ML can discern indicators of attacks from collections of loosely related data faster and more reliably than an overworked (and often under-experienced) analyst.

The image above has been cropped. Could cervical cancer be brought under control? Not quite, but the results of a study published in the Journal of the National Cancer Institute seem promising. Researchers from the National Institutes of Health and Global Good have developed a deep learning algorithm that can analyze digital images of a woman's cervix and identify precancerous changes that require medical attention -- with more accuracy than human experts. The team used comprehensive datasets to train the algorithm to recognize patterns in complex visual inputs, like medical images.