Privacy-Focused AI Platform Dark Unveiled by Eternal
Eternal has unveiled Dark, a privacy-focused AI platform aimed at founders, researchers, scientists, enterprises and governments. Founder Deepinder Goyal said he hopes to launch the platform globally within 2026.
Dark has been in internal use for a few months and uses fine-tuned open-weight models running on Eternal’s infrastructure. It includes end-to-end encryption and an encrypted memory management layer controlled by users, allowing it to retain context across sessions.
With Dark, Eternal is commercialising AI capabilities developed for its own operations. Its approach adapts existing models whose trained parameters are available for developers to use and modify, rather than building a model from scratch.
Newsletter
The corridor, every morning.
Funding rounds and cross-border capital moves, one email, five minutes.
Dark’s Privacy Features
Dark aims to address concerns that conversations with mainstream AI assistants may be retained, reviewed by humans, placed under legal holds or logged by third parties. Goyal argued that these concerns can discourage people from using AI for sensitive or commercially valuable work, including research, health, financial and relationship questions.
Goyal said the platform’s privacy features had encouraged him to share more information and explore questions he would hesitate to ask other AI assistants. “We cannot read your conversations, even if we are forced to,” he said.
More in Startup News
