"We're entering an era in which our enemies can make anyone say anything at any point in time." In this viral video from 2018, actor-writer Jordan Peele projected his voice into former President Obama's moving lips. Peele's PSA on'deepfakes,' audio and video altered with the intent to mislead, was the first time many people heard of synthetic data. It won't be the last. Today, synthetic data are everywhere, driving some of AI's most innovative applications.

Not only did a classifier pre-trained on Task2Sim's fake images perform as well as a model trained on real ImageNet photos, it also outperformed a rival trained on images generated with random simulation parameters. Task2Sim even transferred its know-how to entirely new tasks, creating images to teach a classifier how to identify cactuses and hand-drawn numbers. "The more tasks you use during training, the more generalizable the model will be," Feris said. A related tool, SimVQA,2 also appearing at CVPR, generates synthetic text and images for training robot agents to reason about the visual world. In a typical visual-reasoning task, an agent might be asked to count the number of chairs at a table or identify the color of a bouquet of flowers.

Imagine you're a scientist who needs to discover a new antibiotic to fight off a scary disease. How would you go about finding it? Typically, you'd have to test lots and lots of different molecules in the lab until you find one that has the necessary bacteria-killing properties. You might find some contenders that are good at killing the bacteria only to realize that you can't use them because they also prove toxic to humans. But what if, instead, you could just type into your computer the properties you're looking for and have your computer design the perfect molecule for you? That's the general approach IBM researchers are taking, using an AI system that can automatically generate the design of molecules for new antibiotics.

IBM has unveiled a slew of announcements designed to help businesses scale their use of AI. The company also announced the rollout of new capabilities for its Watson platform. IBM researchers have built a hybrid question-answering system called Neuro-Symbolic-QA (NSQA) that for the first time uses neurosymbolic AI to allow an AI system to offer "and"/ "or" to its recommendations. This will ultimately position the system to perform better in real-world situations, IBM said. "This enhanced reasoning capability comes as a result of an entirely new foundational AI method created by IBM researchers called Logical Neural Networks (LNN), IBM said. LNNs are a modification of today's neural networks so that they become equivalent to a set of logic statements, but they also retain the original learning capability of a neural network, the company explained in a blog post. QA is designed to meet the significant challenges in language-based AI, in particular the fact that the training of NLP ...

IBM this week presented research investigating how AI and machine learning could be used to improve maternal health in developing countries and predict the onset and progression of Type 1 diabetes. In a study funded by the Bill and Melinda Gates Foundation, IBM researchers built models to analyze demographic datasets from African countries, finding "data-supported" links between the number of years between pregnancies and the size of a woman's social network with birth outcomes. In a separate work, a team from IBM analyzed data across three decades and four countries to attempt to anticipate the onset of Type 1 diabetes anywhere from 3 to 12 months before it's typically diagnosed and then predict its progression. They claim one of the models accurately predicted progression 84% of the time. Despite a global decline in child mortality rates, many countries aren't on track to achieving proposed targets of ending preventable deaths among newborns and children under the age of 5. Unsurprisingly, the progress toward these targets remains uneven, reflected in disparities in access to healthcare services and inequitable resource allocation. Toward potential solutions, researchers at IBM attempted to identify features associated with neonatal mortality "as captured in nationally representative cross-sectional data."

A broken computer, appliance or car used to mean a visit to a technician or mechanic, but the recent proliferation of DIY videos has more people performing such repair jobs themselves. Now, a pair of IBM researchers have taken instructional video to the next level, with a new fine-grained visual recognition approach and augmented reality (AR) system that can look at the actual piece of hardware being working on and integrate real-time, step-by-step tech support and guidance. The researchers say the proposed method can increase the rate of first-time fixes and reduce hardware disruption recovery time by automatically detecting the state of an object and presenting the right set of information in the right context. AR basically overlays media and graphics on what we see in the real world. Major technological advances and the increased availability of AR software development kits (SDKs) such as ARKit and ARCore over the last decade have lowered the entry barrier for AR developers. In recent years, machine learning has informed the emergence of intelligent systems that further enhance the AR experience.

Artificial intelligence (AI) systems face a set of conflicting goals: being accurate (consuming large amounts of computational power and electrical power) and being accessible (being lower in cost, less computationally intensive, and less power-hungry). Unfortunately, many of today's AI implementations are environmentally unsustainable. Improvements in AI energy efficiency will be driven by several factors, including more efficient algorithms, more efficient computing architectures, and more efficient components. It's necessary to measure and track the energy consumption of AI systems to identify any improvements in energy efficiency. One example of the increasing awareness of the importance of energy consumption in AI systems is having is reflected in the fact that the ULPMark (ultra-low power) benchmark line from EEMBC is now adding ML inference and developing a new benchmark, the ULPMark-ML.

The field of artificial intelligence (AI) experienced tremendous scientific advances in the last few years, from vast improvements in processing power and computational efficiency to new insights into object identification, language, and deep learning. IBM -- a leader in AI research since its inception in the 1950s -- helped inform many of these advances. And 2019 in particular was a watershed moment for IBM Research AI. Over the course of the year, IBM researchers hit a new record of 175 regular accepted papers at eight of the top AI conferences, hosted the second annual AI Research Week in September, and launched the AI Hardware Center to further exploration of next-generation AI hardware. The MIT-IBM Watson AI Lab, now in its second year, flourished -- welcoming Boston Scientific, Nexplore, Refinitiv and Samsung as inaugural members of its new Membership Program.

Like the IBM researcher who presented on a joint project between IBM and MIT to create smarter AIs, I'm not a fan of the term AI. But don't get me started on naming because, when I did it, I discovered a new rule and that rule is that "the only thing folks will agree on when it comes to a new name is that the poor sap that came up with it is an idiot." Still to me something is either intelligent or not and if you are going to define a class of intelligence you likely should connect it to the source like human intelligence, animal intelligence, or, in this case, Machine Intelligence. Our current AI technology level is stupid. We call it Narrow AI, but it means the AI can do one of a limited number of highly defined tasks somewhat autonomously.

IBM researchers at Black Hat USA 2018 announced their development of DeepLocker, described as a highly targeted and evasive attack tool powered by AI. The new tool is designed to better understand how several existing AI models can be combined with current malware techniques to create a particularly challenging new breed of malware, according to an Aug. 8 blog post. Researchers said the malware is designed to conceal its intent until it reaches a specific victim and then unleashits malicious action as soon as the AI model identifies the target via indicators such as facial recognition, geolocation and voice recognition. To demonstrate the malware's capabilities, researchers designed a proof of concept which camouflaged the WannaCry ransomware in a benign video conferencing application so that it remains undetected by malware analysis tools. The researchers' goal for developing the ransomware was to raise awareness of AI-powered threats, demonstrate how attackers have the capability to build stealthy malware that can circumvent commonly deployed defenses, and provide insights into how to reduce risks and deploy adequate countermeasures.