The machine learning company OpenAI is developing models that improve computer vision and can produce original images from a text prompt. From a report: The new models are the latest steps in ongoing efforts to create machine learning systems that exhibit elements of general intelligence, while performing tasks that are actually useful in the real world -- without breaking the bank on computing power. OpenAI this week is announcing two new systems that attempt to do for images what its landmark GPT-3 model did last year for text generation. DALL-E is a neural network that can "take any text and make an image out of it," says Ilya Sutskever, OpenAI co-founder and chief scientist. That includes concepts it would never have encountered in training, like the drawing of an anthropomorphic daikon radish walking a dog. DALL-E operates somewhat similarly to GPT-3, the huge transformer model that can generate original passages of text based on a short prompt. CLIP, the other new neural network, "can take any set of visual categories and instantly create very strong and reliable visually classifiable text descriptions," says Sutskever, improving on existing computer vision techniques with less training and expensive computational power.

Read more of this story at Slashdot.

Farming equipment like combine harvesters "beam back reams of data to its manufacturer," reports Forbes: GPS records the combine's precise path through the field as it moves. Sensors tally the number of crops gathered per acre and the spacing between them. On a sister machine called a planter, algorithms adjust the distribution of seeds based on which parts of the soil have in past years performed best. Another machine, a sprayer, uses algorithms to scan for weeds and zap them with pesticides. Meanwhile sensors record the wear and tear on the machines, so that when the farmer who operates them heads to the local distributor to look for a replacement part, it has already been ordered and is waiting for them. Farming may be an earthy industry, but much of it now takes place in the cloud. Leading farm machine makers like Chicago-based John Deere or Georgia's AGCO collect data from all around the world thanks to the ability of their bulky machines to extract a huge variety of metrics from farmers' fields and store it online... The amassing of all that data in the hands of the few major companies that sell farm equipment across the country or worldwide has opened up big opportunities for the "smart farming" industry, even as many in the farming community are reluctant to part with information about the fields they plow.... Equipment makers with sufficient sales of machines around the country may in theory actually be able to predict, at least to some small but meaningful extent, the prices of various crops by analyzing the data its machines are sending in — such as "yields" of crops per acre, the amount of fertilizer used, or the average number of seeds of a given crop planted in various regions. Were the company then to sell that data to a commodities trader, say, it could likely reap a windfall: normally, the markets must wait for highly-anticipated government surveys to run their course.

Read more of this story at Slashdot.

Rohit Prasad, Vice President and Head Scientist of Alexa at Amazon, writes: While Turing's original vision continues to be inspiring, interpreting his test as the ultimate mark of AI's progress is limited by the era when it was introduced. For one, the Turing Test all but discounts AI's machine-like attributes of fast computation and information lookup, features that are some of modern AI's most effective. The emphasis on tricking humans means that for an AI to pass Turing's test, it has to inject pauses in responses to questions like, "do you know what is the cube root of 3434756?" or, "how far is Seattle from Boston?" In reality, AI knows these answers instantaneously, and pausing to make its answers sound more human isn't the best use of its skills. Moreover, the Turing Test doesn't take into account AI's increasing ability to use sensors to hear, see, and feel the outside world. Instead, it's limited simply to text. To make AI more useful today, these systems need to accomplish our everyday tasks efficiently. If you're asking your AI assistant to turn off your garage lights, you aren't looking to have a dialogue. Instead, you'd want it to fulfill that request and notify you with a simple acknowledgment, "ok" or "done." Even when you engage in an extensive dialogue with an AI assistant on a trending topic or have a story read to your child, you'd still like to know it is an AI and not a human. In fact, "fooling" users by pretending to be human poses a real risk. Imagine the dystopian possibilities, as we've already begun to see with bots seeding misinformation and the emergence of deep fakes. Instead of obsessing about making AIs indistinguishable from humans, our ambition should be building AIs that augment human intelligence and improve our daily lives in a way that is equitable and inclusive. A worthy underlying goal is for AIs to exhibit human-like attributes of intelligence -- including common sense, self-supervision, and language proficiency -- and combine machine-like efficiency such as fast searches, memory recall, and accomplishing tasks on your behalf. The end result is learning and completing a variety of tasks and adapting to novel situations, far beyond what a regular person can do.

Read more of this story at Slashdot.

schwit1 quotes Intelligent Living: Plenty is an ag-tech startup in San Francisco, co-founded by Nate Storey, that is reinventing farms and farming. Storey, who is also the company's chief science officer, says the future of farms is vertical and indoors because that way, the food can grow anywhere in the world, year-round; and the future of farms employ robots and AI to continually improve the quality of growth for fruits, vegetables, and herbs. Plenty does all these things and uses 95% less water and 99% less land because of it. Plenty's climate-controlled indoor farm has rows of plants growing vertically, hung from the ceiling. There are sun-mimicking LED lights shining on them, robots that move them around, and artificial intelligence (AI) managing all the variables of water, temperature, and light, and continually learning and optimizing how to grow bigger, faster, better crops. These futuristic features ensure every plant grows perfectly year-round. The conditions are so good that the farm produces 400 times more food per acre than an outdoor flat farm. Another perk of vertical farming is locally produced food. The fruits and vegetables aren't grown 1,000 miles away or more from a city; instead, at a warehouse nearby. Meaning, many transportation miles are eliminated, which is useful for reducing millions of tons of yearly CO2 emissions and prices for consumers. Imported fruits and vegetables are more expensive, so society's most impoverished are at an extreme nutritional disadvantage. Vertical farms could solve this problem.

Read more of this story at Slashdot.

DeepMind's latest AI program can attain "superhuman performance" in tasks without needing to be given the rules. From a report: Like the research hub's earlier artificial intelligence agents, MuZero achieved mastery in dozens of old Atari video games, chess, and the Asian board games of Go and Shogi. But unlike its predecessors, it had to work out their rules for itself. It is already being put to practical use to find a new way to encode videos, which could slash YouTube's costs. [...] MuZero could soon be put to practical use too. Dr Silver said DeepMind was already using it to try to invent a new kind of video compression. "If you look at data traffic on the internet, the majority of it is video, so if you can compress video more effectively you can make massive savings," he explained. "And initial experiments with MuZero show you can actually make quite significant gains, which we're quite excited about." He declined to be drawn on when or how Google might put this to use beyond saying more details would be released in the new year. However, as Google owns the world's biggest video-sharing platform -- YouTube -- it has the potential to be a big money-saver. DeepMind is not the first to try and create an agent that both models the dynamics of the environment it is placed in and carries out tree searches -- deciding how to proceed by looking several steps ahead to determine the best outcome. However, previous attempts have struggled to deal with the complexity of "visually rich" challenges, such as those posed by old video games like Ms Pac-Man.

Read more of this story at Slashdot.

Alphabet's Google this year moved to tighten control over its scientists' papers by launching a "sensitive topics" review, and in at least three cases requested authors refrain from casting its technology in a negative light, Reuters reported Wednesday, citing internal communications and interviews with researchers involved in the work. From a report: Google's new review procedure asks that researchers consult with legal, policy and public relations teams before pursuing topics such as face and sentiment analysis and categorizations of race, gender or political affiliation, according to internal webpages explaining the policy. "Advances in technology and the growing complexity of our external environment are increasingly leading to situations where seemingly inoffensive projects raise ethical, reputational, regulatory or legal issues," one of the pages for research staff stated. Reuters could not determine the date of the post, though three current employees said the policy began in June. The "sensitive topics" process adds a round of scrutiny to Google's standard review of papers for pitfalls such as disclosing of trade secrets, eight current and former employees said. For some projects, Google officials have intervened in later stages. A senior Google manager reviewing a study on content recommendation technology shortly before publication this summer told authors to "take great care to strike a positive tone," according to internal correspondence read to Reuters.

Read more of this story at Slashdot.

An anonymous reader quotes a report from Phys.Org: A team of scientists at Freie Universitat Berlin has developed an artificial intelligence (AI) method for calculating the ground state of the Schrodinger equation in quantum chemistry. The goal of quantum chemistry is to predict chemical and physical properties of molecules based solely on the arrangement of their atoms in space, avoiding the need for resource-intensive and time-consuming laboratory experiments. In principle, this can be achieved by solving the Schrodinger equation, but in practice this is extremely difficult. Up to now, it has been impossible to find an exact solution for arbitrary molecules that can be efficiently computed. But the team at Freie Universitat has developed a deep learning method that can achieve an unprecedented combination of accuracy and computational efficiency. The deep neural network designed by [the] team is a new way of representing the wave functions of electrons. "Instead of the standard approach of composing the wave function from relatively simple mathematical components, we designed an artificial neural network capable of learning the complex patterns of how electrons are located around the nuclei," [Professor Frank Noe, who led the team effort] explains. "One peculiar feature of electronic wave functions is their antisymmetry. When two electrons are exchanged, the wave function must change its sign. We had to build this property into the neural network architecture for the approach to work," adds [Dr. Jan Hermann of Freie Universitat Berlin, who designed the key features of the method in the study]. This feature, known as 'Pauli's exclusion principle,' is why the authors called their method 'PauliNet.' Besides the Pauli exclusion principle, electronic wave functions also have other fundamental physical properties, and much of the innovative success of PauliNet is that it integrates these properties into the deep neural network, rather than letting deep learning figure them out by just observing the data. "Building the fundamental physics into the AI is essential for its ability to make meaningful predictions in the field," says Noe. "This is really where scientists can make a substantial contribution to AI, and exactly what my group is focused on." The results were published in the journal Nature Chemistry.

Read more of this story at Slashdot.

Microsoft says in a blog post that PepsiCo is using their Project Bonsai "machine teaching" service to "help ensure its Cheetos cheese-puff snacks all have the same texture, crunch and shape," reports The Wall Street Journal. From the blog post: PepsiCo built a computer vision system that continually monitors Cheeto attributes. Data about qualities such as density and length are fed to the Project Bonsai solution, which makes adjustments to bring the product within spec. This approach reduces the time it takes to correct inconsistencies and allows operators to focus on parts of the line that require human expertise. PepsiCo is preparing to use the solution in a production plant and exploring how to use the solution with other products, including the tortilla chip manufacturing process. An out-of-spec product can't be sold, which leads to wasted resources, time, and money. Greater consistency helps PepsiCo maintain high quality products while maximizing throughput. To make an ideal Cheeto, the solution needed examples of what wasn't ideal -- and needed to know what to do in those cases. The extruder line is self-contained and well-suited for developing and testing an autonomous system solution. Operators had been running it manually, which gave developers the opportunity to build the solution from scratch, instead of on top of other software. The AI solution has a recommendation mode and a closed loop control mode. In both modes, a computer vision system continuously measures the quality of the Cheetos. In recommendation mode, the AI will alert an operator if the product drifts out of spec, displaying on an instrument panel the attributes that are not ideal as well as a recommendation to correct it. The operator can push a button to make any or all recommended adjustments. In control mode, the only difference is that the AI solution skips the recommendation step and adjusts the extruder line specifications independently. The company expects that running this intelligent control system will return product to acceptable attributes faster. In the current extruder line, operators measure product attributes manually at defined intervals. If the Cheetos are out of spec, the operator makes adjustments based on guidelines or experience to return the product to acceptable quality. The problem: Infrequent sampling meant that the line could be producing out-of-spec Cheetos for a longer period of time without anyone realizing. The Project Bonsai solution will monitor the product almost continuously, using sensors to oversee characteristics such as length and bulk density. That way, it knows as soon as the product strays outside a defined range.

Read more of this story at Slashdot.

In Huntsville, Alabama, there is a room with green walls and a green ceiling. Dangling down the center is a fishing line attached to a motor mounted to the ceiling, which moves a procession of guns tied to the translucent line. From a report: The staff at Arcarithm bought each of the 10 best-selling firearm models in the U.S.: Rugers, Glocks, Sig Sauers. Pistols and long guns are dangled from the line. The motor rotates them around the room, helping a camera mounted to a mobile platform photograph them from multiple angles. "It's just like a movie set," said Arcarithm president and CEO Randy E. Riley. This process creates about 5,000 images of each gun floating ethereally. Arcarithm's computer programmers then replace the green backdrop with different environments, like fields, forests, and city streets. They add rain or snow or fog or sun. A program then randomly distorts the images. The result is 30,000 to 50,000 images of the same gun, from multiple angles, in different synthetic settings and of varying degrees of visibility. The point of creating this vast portfolio of digital gun art is to feed an algorithm made to detect a firearm as soon as a security camera catches it being drawn by synthetically creating tens of thousands of ways each gun may appear. Arcarithm is one of several companies developing automated active shooter detection technology in the hopes of selling it to schools, hotels, entertainment venues and the owners of any location that could be the site of one of America's 15,000 annual gun murders and 29,000 gun injuries. Among the other sellers are Omnilert, a longtime vendor of safety notification software, and newcomers ZeroEyes, Defendry, and Athena Securities. Some cities employ a surveillance system of acoustic sensors to instantly detect gunshots. These companies promise to do one better and save precious minutes by alerting police or security personnel before the first shot is fired.

Read more of this story at Slashdot.

Alphabet's Google waived a 1.1 billion-pound loan ($1.5 billion) to DeepMind in 2019 after the U.K.-based artificial intelligence lab continued to ramp up the scale of its research and development. From a report: Revenue jumped 158% in 2019, DeepMind said in a financial filing this week. Sales were 265.5 million pounds, up from 102.8 million pounds a year earlier. Its losses also widened, increasing 1.4% to 476.6 million pounds. DeepMind's parent has agreed to continue funding the company for at least a year after the report's approval. Alphabet's Google Ireland unit waived repayments and interest from the loan to help cover DeepMind's losses. Google acquired DeepMind in 2014 in a 400 million-pound acquisition that gave the Silicon Valley search giant access to cutting edge AI research. DeepMind Chief Executive Officer Demis Hassabis's goal is to produce general-purpose intelligence that can solve an array of problems. It develops products used by its parent company -- like its system for making data centers more energy efficient and a program to improve the accuracy of travel times on Google Maps -- as well as AI with broader applications.

Read more of this story at Slashdot.

On December 15, the United States Air Force successfully flew an AI copilot on a U-2 spy plane in California, marking the first time AI has controlled a U.S. military system. Dr. Will Roper, the Assistant Secretary of the Air Force for Acquisition, Technology and Logistics, reveals how he and his team made history: With call sign ARTUu, we trained uZero -- a world-leading computer program that dominates chess, Go, and even video games without prior knowledge of their rules -- to operate a U-2 spy plane. Though lacking those lively beeps and squeaks, ARTUu surpassed its motion picture namesake in one distinctive feature: it was the mission commander, the final decision authority on the human-machine team. And given the high stakes of global AI, surpassing science fiction must become our military norm. Our demo flew a reconnaissance mission during a simulated missile strike at Beale Air Force Base on Tuesday. ARTUu searched for enemy launchers while our pilot searched for threatening aircraft, both sharing the U-2's radar. With no pilot override, ARTUu made final calls on devoting the radar to missile hunting versus self-protection. Luke Skywalker certainly never took such orders from his X-Wing sidekick! The fact ARTUu was in command was less about any particular mission than how completely our military must embrace AI to maintain the battlefield decision advantage. Unlike Han Solo's "never-tell-me-the-odds" snub of C-3PO's asteroid field survival rate (approximately 3,720 to 1), our warfighters need to know the odds in dizzyingly-complex combat scenarios. Teaming with trusted AI across all facets of conflict -- even occasionally putting it in charge -- could tip those odds in our favor. But to trust AI, software design is key. Like a breaker box for code, the U-2 gave ARTUu complete radar control while "switching off" access to other subsystems. Had the scenario been navigating an asteroid field -- or more likely field of enemy radars -- those "on-off" switches could adjust. The design allows operators to choose what AI won't do to accept the operational risk of what it will. Creating this software breaker box -- instead of Pandora's -- has been an Air Force journey of more than a few parsecs...

Read more of this story at Slashdot.

Amazon today rolled out Live Translation, a new Alexa feature that aims to assist with conversations between people who speak two different languages by leveraging speech recognition and machine translation technology. Amazon says that Live Translation can interpret between a number of dialects in real time, including English and French, Spanish, Hindi, Brazilian Portuguese, German, or Italian. From a report: The pandemic appears to have supercharged voice app usage, which was already on an upswing. According to a study by NPR and Edison Research, the percentage of voice-enabled device owners who use commands at least once a day rose between the beginning of 2020 and the start of April. Just over a third of smart speaker owners say they listen to more music, entertainment, and news from their devices than they did before, and owners report requesting an average of 10.8 tasks per week from their assistant this year compared with 9.4 different tasks in 2019. And according to a new report from Juniper Research, consumers will interact with voice assistants on 8.4 billion devices by 2024. Launching Live Translation requires asking Alexa on an Amazon Echo device to translate one of the supported languages. The command "Alexa, translate French" will translate between English and French, for example, while "Alexa, stop" will end the translation session. The Echo will beep during the session to indicate when to speak in the other language, and Echo devices with a screen like the Echo Show will display a transcription of the conversation. Users can take pauses between sentences, and Alexa will automatically detect the language in which they're speaking and translate each side of the conversation.

Read more of this story at Slashdot.

"Don a headset which places a sensor on the back of your head, and it'll detect your brainwaves which can then be translated into digital actions," writes Engadget. VentureBeat reports that NextMind "has started shipping its real-time brain computer interface Dev Kit for $399." The device translates brain signals into digital commands, allowing you to control computers, AR/VR headsets, and IoT devices (lights, TVs, music, games, and so on) with your visual attention. Paris-based NextMind is part of a growing number of startups building neural interfaces that rely on machine learning algorithms. There are invasive devices like the one from Elon Musk's Neuralink, which in August revealed a prototype showing readings from a pig's brain using a coin-shaped device implanted under the skull. There are also noninvasive devices like the electromyography wristband that translates neuromuscular signals into machine-interpretable commands from Ctrl-labs, which Facebook acquired in September 2019. NextMind is developing a noninvasive device — an electroencephalogram (EEG) worn on the back of your head, where your brain's visual cortex is located. When we spoke with NextMind CEO Sid Kouider last year, he promised the kits would begin shipping in Q2 2020. Then the pandemic hit. "We had about three, four months of delays due to COVID-19, but not more than that in terms of production," Kouider told VentureBeat. The company shipped "hundreds" of Dev Kits in November after producing its first thousand units. Another thousand units are set to be produced next month.

Read more of this story at Slashdot.

An anonymous reader shares a report: More recent Nvidia graphics cards have a proprietary feature called Deep Learning Super Sampling (DLSS), and while it's often been touted as a powerful new rendering tool, the results have sometimes been underwhelming. Some of this is down to the oddly mixed-message around how DLSS was rolled-out: it only works on more recent Nvidia cards that are still near the cutting edge of PC graphics hardware⦠but DLSS is designed to render images at lower resolutions but display them as if they were rendered natively at a higher resolution. If you had just gotten a new Nvidia card and were excited to see what kind of framerates and detail levels it could sustain, what DLSS actually did sounded counterintuitive. Even games like Control, whose support of DLSS was especially praised, left me scratching my head about why I would want to use the feature. On my 4K TV, Control looked and ran identically well with and without DLSS, so why wouldn't I just max-out my native graphics settings instead rather than use a fancy upscaler? Intellectually, I understood the DLSS could produce similarly great looking images without taxing my hardware as much, but I neither fully believed it, nor had I seen a game where the performance gain was meaningful. Cyberpunk 2077 converted me. DLSS is a miracle, and without it there's probably no way I would ever have been happy with my graphics settings or the game's performance. I have a pretty powerful video card, an RTX 2080 TI, but my CPU is an old i5 overclocked to about 3.9 GHz and it's a definite bottleneck on a lot of games. Without DLSS, Cyberpunk 2077 was very hard to get running smoothly. The busiest street scenes would look fine if I were in a static position, but a quick pan with my mouse would cause the whole world to stutter. If I was walking around Night City, I would get routine slow-downs. Likewise, sneaking around and picking off guards during encounters was all well and good but the minute the bullets started flying, with grenades exploding everywhere and positions changing rapidly, my framerate would crater to the point where the game verged on unplayable. To handle these peaks of activity, I had to lower my detail settings way below what I wanted, and what my hardware could support for about 80 percent of my time with the game. Without DLSS, I never found a balance I was totally happy with. The game neither looked particularly great, nor did it run very well. DLSS basically solved this problem for me. With it active, I could run Cyberpunk at max settings, with stable framerates in all but the busiest scenes.

Read more of this story at Slashdot.

Salesforce today peeled back the curtains on ReceptorNet, a machine learning system researchers at the company developed in partnership with clinicians at the University of Southern California's Lawrence J. Ellison Institute for Transformative Medicine of USC. From a report: The system, which can determine a critical biomarker for oncologists when deciding on the appropriate treatment for breast cancer patients, achieved 92% accuracy in a study published in the journal Nature Communications. Breast cancer affects more than 2 million women each year, with around one in eight women in the U.S. developing the disease over the course of their lifetime. In 2018 in the U.S. alone, there were also 2,550 new cases of breast cancer in men. And rates of breast cancer are increasing in nearly every region around the world. In an effort to address this, Salesforce researchers developed an algorithm -- the aforementioned ReceptorNet -- that can predict hormone-receptor status from inexpensive and ubiquitous images of tissue. Typically, breast cancer cells extracted during a biopsy or surgery are tested to see if they contain proteins that act as estrogen or progesterone receptors. (When the hormones estrogen and progesterone attach to these receptors, they fuel the cancer growth.) But these types of biopsy images are less widely available and require a pathologist to review. In contrast to the immunohistochemistry process favored by clinicians, which requires a microscope and tends to be expensive and not readily available in parts of the world, ReceptorNet determines hormone receptor status via hematoxylin and eosin (H&E) staining, which takes into account the shape, size, and structure of cells. Salesforce researchers trained the system on several thousand H&E image slides from cancer patients in "dozens" of hospitals around the world.

Read more of this story at Slashdot.

Apple has moved its self-driving car unit under the leadership of top artificial intelligence executive John Giannandrea, who will oversee the company's continued work on an autonomous system that could eventually be used in its own car, Bloomberg reports. From the report: The project, known as Titan, is run day-to-day by Doug Field. His team of hundreds of engineers have moved to Giannandrea's artificial intelligence and machine-learning group, according to people familiar with the change. Previously, Field reported to Bob Mansfield, Apple's former senior vice president of hardware engineering. Mansfield has now fully retired from Apple, leading to Giannandrea taking over. Giannandrea joined Apple in 2018 as its vice president of AI Strategy and Machine Learning before being promoted to Apple's executive team as a senior vice president later that year. He ran Google's machine-learning and search teams before that. At Apple, in addition to the car project, he is in charge of Siri and machine-learning technologies across Apple's products. Mansfield initially retired from Apple in 2012, only to return for less than a year as its senior vice president in charge of chip technology. Mansfield stepped down from that role in 2013 and then remained as a part-time consultant.

Read more of this story at Slashdot.

Google CEO Sundar Pichai apologized Wednesday for the company's handling of the departure of AI ethics researcher Timnit Gebru and said he would investigate the events and work to restore trust, according to an internal memo sent companywide and obtained by Axios. From the report: Gebru's exit has provoked anger and consternation within Google as well as in academic circles, with thousands of people signing an open letter urging Google to reexamine its practices. In the note, Pichai acknowledged the depth of the damage done by the company's actions and said the company would look at all aspects of the situation, but stopped short of saying the company made a mistake in removing Gebru. "I've heard the reaction to Dr. Gebru's departure loud and clear: it seeded doubts and led some in our community to question their place at Google," Pichai said in the memo. " I want to say how sorry I am for that, and I accept the responsibility of working to restore your trust." While Pichai's memo strikes a contrite tone, it's unclear how far it will go to addressing the significant upset within Google's ranks, especially among those concerned with its commitments to diversity and academic freedom.

Read more of this story at Slashdot.

Google today launched an experimental app for Android that leverages AI to make communication more accessible for people with speech and motor impairments. Called Look to Speak, it tracks eye movements to let people use their eyes to select prewritten, customizable phrases and have them spoken aloud. From a report: Approximately 18.5 million people in the U.S. have a speech, voice, or language impairment. Eye gaze-tracking devices can provide a semblance of independence, but they're often not portable and tend to be expensive. The entry-level Tobii 4C eye tracker starts at $150, for instance. To address this need, speech and language therapist Richard Cave began collaborating with a small group at Google to develop Look to Speak. The app, which is available for free and compatible with Android 9.0 and above, enables users to glance left, right, or up to select what they wish to say from a phrase list. With Look to Speak, people can personalize the words and sentences on their list and adjust eye gaze sensitivity settings. Google says the app's data remains private and never leaves the phone on which it's installed.

Read more of this story at Slashdot.

President Donald Trump on Thursday signed an executive order that aims to guide how federal agencies adopt artificial intelligence (AI) as part of efforts to build public trust in the government using this technology. From a report: The order itself directs federal agencies to be guided by nine principles when designing, developing, acquiring, and using AI. These principles emphasise that AI use by federal agencies be lawful; purposeful and performance-driven; accurate, reliable, and effective; safe, secure, and resilient; understandable; responsible and traceable; regularly monitored; transparent; and accountable. To implement these principles, the order directs the Office of Management and Budget to create a roadmap by the end of May 2021 for how the government will better support the use of AI. This roadmap will include a schedule for engaging with the public and timelines for finalising relevant policy guidance. The order also calls on agencies to continue to use voluntary consensus standards developed with industry participation. "This order recognises the potential for AI to improve government operations, such as by reducing outdated or duplicative regulations, enhancing the security of federal information systems, and streamlining application processes," Trump said in a statement. Federal agencies will also be required to prepare an inventory of AI use cases, as well as review and assess these use cases for consistency. The General Services Administration, meanwhile, has been directed to establish an AI track within the Presidential Innovation Fellows program to attract experts from industry and academia to work within agencies to further the design, development, acquisition, and use of AI in government.

Read more of this story at Slashdot.

Timnit Gebru, one of the best-known AI researchers today and co-lead of an AI ethics team at Google, said she was fired Wednesday for sending an email to "non-management employees that is inconsistent with the expectations of a Google manager." VentureBeat reports: She said Google AI employees who report to her were emailed and told that she accepted her resignation when she did not offer her resignation. According to Casey Newton's Platformer, who reportedly obtained a copy, Gebru sent the email in question to the Google Brain Women and Allies listserv. In it, Gebru expresses frustration with the lack of progress in hiring women at Google and lack of accountability for failure to make progress. She also said was told not to publish a piece of research and advised employees to no longer fill out diversity paperwork because it didn't matter. No mention is made of resignation. "There is no way more documents or more conversations will achieve anything. We just had a Black research all hands with such an emotional show of exasperation. Do you know what happened since? Silencing in the most fundamental way possible," the email reads. When asked by VentureBeat for comment, a Google spokesperson provided a link to the Platformer article with a copy of an email sent Thursday by Google AI chief Jeff Dean to company research staff. In it, Dean said a research paper written by Gebru and other researchers was submitted for publication at a conference before completing a review process and addressing feedback. In response, Dean said he received an email from Gebru. "Timnit wrote that if we didn't meet these demands, she would leave Google and work on an end date. We accept and respect her decision to resign from Google," he said. "Given Timnit's role as a respected researcher and a manager in our Ethical AI team, I feel badly that Timnit has gotten to a place where she feels this way about the work we're doing. I also feel badly that hundreds of you received an email just this week from Timnit telling you to stop work on critical DEI programs. Please don't. I understand the frustration about the pace of progress, but we have important work ahead and we need to keep at it."

Read more of this story at Slashdot.