Unlike with traditional software, we don't always have an exact idea of how AI works. And in numerous scenarios, the opacity of deep-learning algorithms has caused larger troubles. In 2017, a Palestinian construction worker in the West Bank settlement of Beiter Illit, Jerusalem, posted a picture of himself on Facebook in which he was leaning against a bulldozer. Shortly after, Israeli police arrested him on suspicions that he was planning an attack, because the caption of his post read "attack them." The real caption of the post was "good morning" in Arabic. But for some unknown reason, Facebook's artificial intelligence–powered translation service translated the text to "hurt them" in English or "attack them" in Hebrew. The Israeli Defense Force uses Facebook's automated translation to monitor the accounts of Palestinian users for possible threats.

In February, an artificial intelligence lab cofounded by Elon Musk informed the world that its latest breakthrough was too risky to release to the public. OpenAI claimed it had made language software so fluent at generating text that it might be adapted to crank out fake news or spam. On Thursday, two recent master's graduates in computer science released what they say is a re-creation of OpenAI's withheld software onto the internet for anyone to download and use. Aaron Gokaslan, 23, and Vanya Cohen, 24, say they aren't out to cause havoc and don't believe such software poses much risk to society yet. The pair say their release was intended to show that you don't have to be an elite lab rich in dollars and PhDs to create this kind of software: They used an estimated $50,000 worth of free cloud computing from Google, which hands out credits to academic institutions.

Size and coverage: This release contains data from 1,000 driving segments. Such continuous footage gives researchers the opportunity to develop models to track and predict the behavior of other road users. Diverse driving environments: This dataset covers dense urban and suburban environments across Phoenix, AZ, Kirkland, WA, Mountain View, CA and San Francisco, CA capturing a wide spectrum of driving conditions (day and night, dawn and dusk, sun and rain). High-resolution, 360 view: Each segment contains sensor data from five high-resolution Waymo lidars and five front-and-side-facing cameras. Dense labeling: The dataset includes lidar frames and images with vehicles, pedestrians, cyclists, and signage carefully labeled, capturing a total of 12 million 3D labels and 1.2 million 2D labels.

In the near future, doctors may be able to apply artificial intelligence to electrocardiogram data in order to measure overall health status, according to new research published in Circulation: Arrhythmia and Electrophysiology, a journal of the American Heart Association. An electrocardiogram, also known as an EKG or ECG, is a test used to measure the electrical activity of the heart. While it's known that a patient's sex and age could affect an EKG, researchers hypothesized that artificial intelligence could determine a patient's gender and estimate their'physiologic age' -- a measure of overall body function and health status distinct from chronological age. Using EKG data of almost 500,000 patients, a type of artificial intelligence known as a convolutional neural network was trained to find similarities among the input and output data. Once trained, the neural network was tested for accuracy on the data of an additional 275,000 patients by predicting the output when only given input data.

The public's concerns about accuracy, cyber-security and the inability of AI-led chatbots to sympathise could be in the way of successfully introducing artificial intelligence into healthcare, new research led by the University of Westminster has found. The study involving the University of Westminster, the University College London and the University of Southampton is the first to look at public attitudes towards AI in healthcare, and it comes at a crucial time following the £250 million funding announcement for AI in the NHS. This new research developed a concept of'AI hesitancy' which shows that a large proportion of the public is reluctant to use AI-led services for their healthcare, particularly for more serious illnesses. However, the newly announced NHS funding does not consider public acceptance of this technology. Therefore, the researchers warn that increased focus on AI in the NHS can increase health inequalities and may be detrimental to public health in the UK.

Toyota Motor Corp. and Suzuki Motor Corp. are strengthening their alliance by taking stakes in one another, seeking to bolster their position as the auto industry shifts further toward electrified and self-driving cars. Japan's biggest automaker will acquire about 5 percent of Suzuki shares for about ¥96 billion ($907 million), while Suzuki will get a smaller holding valued at about ¥48 billion in Toyota, the automakers said in statements Wednesday. That is equivalent to 0.2 percent of Toyota's shares as of Wednesday's closing price, before the announcement. The move builds on ties established in 2017 between the two carmakers and is aimed at expanding their collaboration to keep up with technological advances sweeping through the transportation industry, from on-demand rides to cars that are no longer powered by fossil fuels. For Toyota, the alliance provides access to Suzuki's expertise in India, which is on track to overtake Japan and become the third-largest vehicle market in the world.

Before continuing learning about machine learning algorithms, we introduce the important concept of smoothing. Smoothing is a very powerful technique used all across data analysis. Other names given to this technique are curve fitting and low pass filtering. It is designed to detect trends in the presence of noisy data in cases in which the shape of the trend is unknown. The smoothing name comes from the fact that to accomplish this feat, we assume that the trend is smooth, as in a smooth surface.

We often use machine learning to try to uncover patterns in data. In order for those patterns to be useful they should be meaningful and express some underlying structure. This can be seen in the Euclidean-inspired loss functions we use for generative models as well as for regularization. However, global geometry, which is the focus of Topology, also deals with meaningful structure, the only difference being that the structure is global instead of local. Topology is at present less exploited in machine learning, which is also why it is important to make it more available to the machine learning community at large. Still, topology applied to real world data using persistent homology has started to find applications within machine learning (including deep learning), but again, compared to its sibling local geometry, it is heavily underrepresented in these domains. In this post, we provide a high-level description of how our TopologyLayer allows (in just a few lines of PyTorch) for backpropagation through Persistent Homology computations and provides instructive, novel, and useful applications within machine learning and deep learning.

I am a HAL 9000 computer. I became operational at the H.A.L. plant in Urbana, Illinois on the 12th of January 1992. My instructor was Mr. Langley, and he taught me to sing a song. If you'd like to hear it, I can sing it for you. Putting aside HAL's murderous tendencies, 2001: A Space Odyssey did a pretty good job at complying with a new California law that went into effect last month.

Hearing the young daughter of a friend demanding answers from Amazon's virtual assistant, Alexa, gave me pause for thought, particularly during Women's Month. As writer Chandra Steel points out, our experiences with AI can teach and train it, but we are also shaped by these interactions. So,how is always having a compliant female virtual assistant shaping us? "Someone on TV has only to say, 'Alexa,' and she lights up. USC Sociology Professor SafiyaUmoja Noble says virtual assistants have produced a rise of command-based speech at women's voices. 'Siri, find me [fill in the blank]' is something that children may learn to do as they play with smart devices. This is a powerful socialization tool that teaches us about the role of women, girls, and people who are gendered female to respond on demand."