• Humans Unable to Detect Over a Quarter of Deepfake Speech Samples

    New research has found that humans were only able to detect artificially generated speech 73% of the time, with the same accuracy in both English and Mandarin.

  • Reached: Milestone in Power Grid Optimization on World’s First Exascale Supercomputer

    Ensuring the nation’s electrical power grid can function with limited disruptions in the event of a natural disaster, catastrophic weather or a manmade attack is a key national security challenge. Compounding the challenge of grid management is the increasing amount of renewable energy sources such as solar and wind that are continually added to the grid, and the fact that solar panels and other means of distributed power generation are hidden to grid operators.

  • Aging Bridge Detection Through Digital Image Correlation

    Researchers have developed a novel and practical method of assessing the mechanical properties of structures, with potential application to structural health monitoring of large structures such as bridges and viaducts.

  • Using Artificial Mussels to Monitor Radioactivity in the Ocean

    Amid the global concern over the pollution of radioactive wastes in the ocean, researchers have conducted a study which has found that “artificial mussels” (AMs) can effectively measure low concentrations of radionuclides in the sea. It is believed that this technology can be applied as a reliable and effective solution for monitoring radioactive contamination around the world.

  • Denying Denial-of-Service: Strengthening Defenses Against Common Cyberattack

    A Denial-of-Service attack is a cyberattack that makes a computer or other device unavailable to its intended users. This is usually accomplished by overwhelming the targeted machine with requests until normal traffic can no longer be processed. Scientists have developed a better way to recognize a common internet attack, improving detection by 90 percent compared to current methods.

  • Fighting Fake “Facts” with Two Little Words: Grounding a Large Language Model's Answers in Reality

    Asking ChatGPT for answers comes with a risk—it may offer you entirely made-up “facts” that sound legitimate. Despite having been trained on vast amounts of factual data, large language models, or LLMs, are prone to generating false information called hallucinations. Inspired by a phrase commonly used in journalism, the researchers conducted a study on the impact of incorporating the words “according to” in LLM queries.

  • Fact-Checking Found to Influence Recommender Algorithms

    Researchers have shown that urging individuals to actively participate in the news they consume can reduce the spread of these kinds of falsehoods. “We don’t have to think of ourselves as captive to tech platforms and algorithms,” said a researcher.

  • Fighting Fake News: Using Machine Learning, Blockchain to Counter Misinformation

    False information can lead to harmful consequences. How can content creators focus their efforts on areas where the misinformation is likely to do the most public harm? Research offers possible solutions through a proposed machine learning framework, as well as expanded use of blockchain technology.

  • Using AI to Protect Against AI Image Manipulation

    As we enter a new era where technologies powered by artificial intelligence can craft and manipulate images with a precision that blurs the line between reality and fabrication, the specter of misuse looms large. “PhotoGuard,” developed by MIT CSAIL researchers, prevents unauthorized image manipulation, safeguarding authenticity in the era of advanced generative models.

  • New Cipher System Protects Computers Against Spy Programs

    Researchers have achieved a breakthrough in computer security with the development of a new and highly efficient cipher for cache randomization. The innovative cipher addresses the threat of cache side-channel attacks, offering enhanced security and exceptional performance.

  • De-Risking Authoritarian AI

    You may not be interested in artificial intelligence, but it is interested in you. AI-enabled systems make many invisible decisions affecting our health, safety and wealth. They shape what we see, think, feel and choose, they calculate our access to financial benefits as well as our transgressions. In a technology-enabled world, opportunities for remote, large-scale foreign interference, espionage and sabotage —via internet and software updates—exist at a ‘scale and reach that is unprecedented’.

  • Regulate National Security AI Like Covert Action

    Congress is trying to roll up its sleeves and get to work on artificial intelligence (AI) regulation. Ashley Deeks writes that only a few of these proposed provisions, however, implicate national security-related AI, and none create any kind of framework regulation for such tools. She proposes crafting a law similar to the War Powers Act to govern U.S. intelligence and military agencies use of AI tools.

  • Bringing Resilience to Small-Town Hydropower

    Using newly developed technologies, researchers demonstrated how hydropower with advanced controls and use of a mobile microgrid, can enable small communities to maintain critical services during emergencies.

  • U.S. Voluntary AI Code of Conduct and Implications for Military Use

    Seven technology companies including Microsoft, OpenAI, Anthropic and Meta, with major artificial intelligence (AI) products made voluntary commitments regarding the regulation of AI. These are non-binding, unenforceable and voluntary, but they may form the basis for a future Executive Order on AI, which will become critical given the increasing military use of AI.

  • Geoscientists Aim to Improve Human Security Through Planet-Scale POI Modeling

    Geoinformatics engineering researchers developed MapSpace, a publicly available, scalable land-use modeling framework. By providing data characteristics broader and deeper than satellite imagery alone, MapSpace can generate population analytics invaluable for urban planning and disaster response.