ROME, N.Y. – U.S. military researchers are asking Lockheed Martin Corp. to continue work on prototyping a system to detect and defeat automated enemy disinformation campaigns launched by manipulating the Internet, news, and entertainment media.
Officials of the Air Force Research Laboratory Information Directorate in Rome, N.Y., announced a $19.3 million order in August to the Lockheed Martin Advanced Technology Laboratories in Cherry Hill, N.J., to finish a prototype for the Semantic Forensics (SemaFor) program.
SemaFor is developing technologies to detect, attribute, and characterize falsified multi-modal media like text, audio, images, and video automatically to defend against large-scale automated disinformation attacks.
The Air Force Research Lab awarded the contract to Lockheed Martin on behalf of the U.S. Defense Advanced research Projects Agency (DARPA) in Arlington, Va.
Statistical detection techniques have been successful, yet media generation and manipulation technology is advancing rapidly. Purely statistical detection methods quickly are becoming insufficient for detecting falsified media.
Detection techniques that rely on statistical fingerprints, moreover, often can be fooled with limited additional resources like algorithm development, data, or computers.
Yet existing automated media manipulation and generation algorithms rely heavily on purely data driven approaches and are prone to making semantic errors. Faces generated by the Generative Adversarial Network (GAN), for example, may have semantic inconsistencies such as mismatched earrings, which provide an opportunity for defenders to gain an asymmetric advantage.
A suite of semantic inconsistency detectors would increase the burden on media falsifiers by requiring the creators of falsified media to get every semantic detail correct, while defenders only need to find one, or a very few, inconsistencies.
SemaFor seeks to develop semantic technologies for analyzing media. Semantic detection algorithms will determine if media is generated or manipulated. Attribution algorithms will infer if media originates from a particular organization or individual. Characterization algorithms will reason about whether media was generated or manipulated for malicious purposes.
The results of detection, attribution, and characterization algorithms can help develop explanations for system decisions, and rank assets for analyst review. These SemaFor technologies will help identify, deter, and understand adversary disinformation campaigns.
DARPA announced initial SemaFor contracts in 2021 to teams led by Kitware, Inc., Purdue University, SRI International, and the University of California at Berkeley.
SemaFor is developing technologies for automatically assembling and curating the evidence provided by the detection, attribution, and characterization algorithms. Lockheed Martin led the research team taking on development of these technologies and has been developing a prototype SemaFor system.
This order brings the total cumulative face value of the Lockheed Martin contract to $37.2 million. On this order, Lockheed Martin will do the work in Cherry Hill, N.J., and should be finished by October 2026.
For more information contact Lockheed Martin Advanced Technology Laboratories online at www.lockheedmartin.com/en-us/capabilities/research-labs/advanced-technology-labs.html, or DARPA at www.darpa.mil/program/semantic-forensics.
