Livepeer AI tackles one of the most pressing challenges of the AI boom: the exponential demand for AI computing capacity
Open video infrastructure network Livepeer has announced the launch of its AI subnet. As the use of generative AI increases, there is a commensurate need for more customizable, dynamic access to AI computing capabilities for builders, makers and developers, according to Livepeer CEO Doug Petkanics : “AI inference is one of the most expensive and compute-intensive challenges in generative AI,” said Livepeer CEO Doug Petkanics. “Our global network of decentralized nodes can meet the growing demand for AI computing – a demand that will only grow exponentially as text-to-video models become embedded in entertainment, social media and gaming.”
The Livepeer AI subnet allows hardware vendors to earn fees for contributing their GPUs for AI inference tasks such as text-to-image, text-to-video, and image-to-video. The subnet is a fork of the main Livepeer network, providing new AI capabilities while still using the Livepeer protocol for node discovery and payment.
The goal of launching the subnet is to onboard nodes to support Livepeer’s new AI capabilities, conduct testing and collect data from strategic testing partners that will evaluate the cost-effectiveness of the Livepeer network for running AI benchmarking inference at scale, Petkanics continues:
“Once validated, the team will merge these capabilities across Livepeer’s core customers, add additional job types, and grow the ecosystem around the use of additional forms of AI-based video computing. The goal is to enable AI developers to build their own models, implement weights, fine-tuning or custom LoRAs on top of existing base models within the network. These give the ultimate flexibility to developers, who can design custom workflows to support any conceivable type of AI media compute need.”
Also Read: VPS AI Announces Its Launch – Pioneering Decentralized Cloud Computing Solutions
The team is currently targeting an August launch of the Livepeer AI mainnet, allowing builders, makers and developers to leverage Livepeer’s cost-efficient, reliable and scalable infrastructure for all their generative AI computing needs.
Livepeer joins the C2PA in the fight against deepfakes
Livepeer is the first open infrastructure network to join the Coalition for Content Provenance and Authenticity ( C2PA ), an open technical standard that gives publishers, creators and consumers the ability to trace the provenance of different types of media. Other members include TikTok, Adobe, Google, Sony, Intel, BBC, Microsoft and OpenAI.
“In AI, content provenance is difficult,” says Petkanics. “The infrastructure layer is ideally positioned to provide a critical stopgap to misinformation, as it can seamlessly fingerprint content as it is created, edited, distributed and consumed. We are committed to reducing the support content authenticity to rebuild trust and embed verifiable metadata that enables attribution at every stage of the content lifecycle.”
Source: PRNewsWire
