More on the topic...
Generating detailed summary...
Failed to generate summary. Please try again.
Higgsfield rolled out Supercomputer 2.0, an autonomous marketing agent built on NVIDIA’s Agent Toolkit and powered by Nemotron and proprietary Soul models. The startup says 78 percent of Fortune 500 firms use its platform to handle the entire campaign lifecycle—from brainstorming to posting and optimization—though there’s no third-party audit of that figure. The system ties together more than 35 image, audio and video models and offers over 20 production pipelines for everything from TV spots to Amazon product reels.
Under the hood, Supercomputer 2.0 adds policy guardrails to catch data leaks, fine-grained permission controls and an audit layer for compliance teams—though the audit functionality won’t ship at launch. Higgsfield claims its net revenue nearly quadrupled in the first five months of 2026, driven by a steady 30 percent month-over-month increase, and reports 12,000 businesses using its tools across six continents.
The company has raised about $138 million so far, including an $80 million Series A extension led by Accel in January 2026. Founder Alex Mashrabov, formerly head of generative AI at Snap, led a 15-person team that used Supercomputer 2.0 to produce Hell Grind, a 95-minute AI-generated action fantasy film, in 14 days for under $500,000. That film premiered at Cannes, where critics noted its surprisingly realistic visuals as traditional studios worry about AI’s growing role.
Higgsfield faces competition from Zyg, Gradial and NeoCognition, and from Meta’s Advantage+ suite, which already serves 8 million advertisers. McKinsey figures suggest agentic AI could handle up to two-thirds of marketing work and boost campaign speed by 15×, but fewer than 10 percent of CMOs have deployed full end-to-end workflows. Higgsfield argues its safety and permission features close that gap.
Questions about this article
No questions yet.