Reality Defender, a startup developing tools for detecting deepfakes and other content created by artificial intelligence, announced that it managed to raise investments for $15 million.
The funds were received by the specified company following the results of the Series A financing round, which was headed by DCVC. Comcast Ventures, Ex/ante, Parameter Ventures, and Nat Friedman’s AI Grant also participated in this process.
The co-founder and CEO of the startup, Ben Colman, said that the funds received will be used to double the number of employees of the firm, which currently stands at 23 people. The company will also spend part of the investment on improving existing models of detection content, which was created using machine intelligence.
According to the head of the startup, the process of the emergence of new methods of generating information and other materials is constant, and the next form of forgery will take the world by surprise when it appears. Ben Colman noted that in this case, people will be shocked both by the quality of the falsification in terms of its compliance with what can be called reliable, and by the scale of damage caused as a result of the use of artificial intelligence technologies in scenarios that go beyond the space of legitimate and responsible apply of AI.
Reality Defender adheres to a research approach in the fight against deepfakes. This conceptual framework of activity allows the company to outpace new generations of methods and models for creating content that can be characterized as falsified. In this case, it is assumed that the startup identifies new technologies before the appearance of the deepfakes created with its help in the public space.
Ben Colman, former vice president of Goldman Sachs, launched Reality Defender in 2021 with Ali Shahriyari and Gaurav Bharaj. At the initial stage of its activity, the company positioned itself as a non-profit organization. Ben Colman said that the startup decided to raise external funding after a final understanding of the wide scale of the spread of deepfakes was formed and there was an increase in demand for technologies that allow detecting materials that were created by machine intelligence and do not reflect the real state of affairs in a particular sphere.
Speaking about the problem of false content, the head of the company does not exaggerate to advertise own solutions and services. DeepMedia, a competitor of Reality Defender, estimated that since the beginning of this year, three times more fake videos and audio recordings have been recorded in the Internet space than in the same period of 2022. This growth trend is partly explained by the scaling of such processes as the commercialization of generative artificial intelligence tools.
Cloning a voice or creating a deep forgery of an image or video a few years ago involved costs of several thousand dollars and access to advanced technological solutions. Modern platforms, such as ElevenLabs, which falsifies audio information, and open source models like Stable Diffusion, which generates images, allow to creation of content that forms an idea of reality in the consumer’s mind that does not correspond to what is actually happening in the surrounding space, are offered almost free of charge.
As we have reported earlier, Paxton AI Raises $6 Million for Legal Research and Drafting Platform.