Stability AI

Open-Source Creative AI Platform

Founded: 2020 | HQ: London, United Kingdom | Key People: Prem Akkaraju (CEO), Sean Parker (Executive Chairman), Hanno Basse (CTO), Robin Rombach (Lead Scientist) | Valuation: ~$1 billion | Flagship Models: Stable Diffusion 3.5 Large, SD3.5 Medium, Stable Audio 3.0, Stable Image Ultra, Stable Video 4D 2.0, Stable Fast 3D

The Origin Story

Stability AI's story is inseparable from its founder Emad Mostaque, a charismatic and controversial figure whose grand vision for democratizing AI image generation collided spectacularly with the realities of running a venture-backed company. Stability AI did not create Stable Diffusion — the model was developed by researchers at Ludwig Maximilian University of Munich and Runway, with compute provided by Stability AI. What Mostaque did was commercialize and popularize it, releasing Stable Diffusion as open-source software in August 2022 and positioning Stability AI as the torchbearer for open, accessible generative AI. The model's ability to run locally on consumer hardware, without API fees or content restrictions, ignited a global creative revolution and made Stability AI an instant darling of the open-source AI movement. A $101 million funding round in October 2022 valued the company at $1 billion, with Coatue, Lightspeed Venture Partners, and O'Shaughnessy Ventures participating. Mostaque painted an ambitious vision of open AI models across every modality — images, language, audio, video, and 3D — funded by enterprise subscriptions and consulting services.

Key Milestones

Stable Diffusion's open-source release in August 2022 was a watershed moment in AI history, enabling thousands of derivative products, services, and artistic projects. By 2023, Stable Diffusion had been downloaded hundreds of millions of times, and Stability AI claimed its models powered roughly 80% of all AI-generated images worldwide. The company expanded into language models with StableLM and into audio with Stable Audio. Stable Diffusion XL (SDXL) in July 2023 improved image quality significantly. However, the company's internal trajectory was increasingly turbulent. A devastating Forbes investigation in June 2023 revealed that Mostaque had made misleading claims about his academic credentials (he did not receive a master's degree from Oxford), exaggerated partnerships with Amazon (which was a standard cloud computing arrangement, not a strategic partnership as described to investors), and fabricated relationships with the OECD, WHO, and World Bank. Co-founder Cyrus Hodes sued Mostaque, alleging he was duped into selling his 15% stake for $100 just weeks before the $101 million funding round. The company was burning approximately $8 million per month by October 2023 and had failed to raise additional funding at a $4 billion target valuation. Multiple senior executives departed amid reports of unpaid bills to cloud providers, including Amazon threatening to revoke GPU access over millions in unpaid invoices. In March 2024, Mostaque resigned as CEO and from the board, stating that “you're not going to beat centralized AI with more centralized AI” and leaving to pursue decentralized AI projects. In June 2024, former Facebook president Sean Parker led an $80 million investment, becoming executive chairman, and Prem Akkaraju was appointed CEO. The company released Stable Diffusion 3.5 in late 2024, stabilizing its product roadmap. Filmmaker James Cameron joined the board of directors, signaling ambition in AI-generated video content. By 2025, the company had approximately 270 million model downloads and was focusing on commercializing its technology for enterprise media and entertainment use cases. The recovery accelerated through 2025: the company erased all remaining debt and achieved triple-digit revenue growth, earning a 2026 Global Recognition Award for “the most complete AI startup recovery in industry history.” Hanno Basse, former CTO of Weta Digital, joined as Chief Technology Officer. Three of Stability AI's text-to-image models (Stable Image Ultra, Stable Diffusion 3 Large, and Stable Image Core) were made available on Amazon Bedrock, and Stable Diffusion 3.5 Large launched on Amazon SageMaker JumpStart. The product line expanded to include Stable Virtual Camera (2D-to-3D from a single image), Stable Video 4D 2.0, Stable Audio Open (Arm mobile-optimized for on-device inference), and Stable Fast 3D, establishing Stability AI as the only enterprise platform offering open-weight models across all five creative AI modalities — image, video, audio, 3D, and code.

Stable Audio 3.0 and Music AI Push

On May 20, 2026, Stability AI released Stable Audio 3.0, a four-model family capable of generating up to six minutes of audio per prompt. Three of the four models were released as open-weight on HuggingFace, continuing Stability AI's commitment to open-source accessibility while reserving the highest-tier model for commercial licensing. The release positions Stability AI as a serious contender in AI music and audio generation, a market that has attracted growing investment from both technology companies and the music industry. To lead its professional music offering, Stability AI brought on Ethan Kaplan, a veteran of Universal Audio and Fender, signaling intent to build tools for professional musicians and audio engineers rather than just casual creators.

Brand Studio and Enterprise Creative Platform

On April 8, 2026, Stability AI launched Brand Studio, an enterprise creative production platform that enables brands to generate consistent, on-brand visual content at scale using Stability AI's image and video models. The platform addresses a key enterprise pain point — maintaining brand consistency while leveraging generative AI — and represents Stability AI's most ambitious push into commercial creative tools. Brand Studio targets marketing departments, advertising agencies, and media companies that need high-volume visual content production with brand governance controls.

Legal Challenges: Getty Images

The Getty Images lawsuit against Stability AI reached a significant milestone when the UK High Court confirmed that Stability AI's models had infringed Getty's trademarks. The ruling, while limited to the trademark claim and not a final judgment on all copyright issues, establishes legal precedent for AI-generated content that incorporates protected brand assets. The case continues to have implications for the broader generative AI industry, as courts worldwide grapple with how intellectual property law applies to AI models trained on internet-scraped data.

Current Position

Stability AI is executing a multi-pronged recovery and expansion strategy under the Parker-Akkaraju leadership. The company maintains its position as the only enterprise platform offering open-weight models across all five creative AI modalities — image, video, audio, 3D, and code — while building commercial products like Brand Studio that generate revenue beyond raw model access. Stable Audio 3.0 and the hiring of Ethan Kaplan signal serious intent in professional audio, a relatively less contested market than image generation. The Getty Images ruling introduces legal risk, though the precedent affects the broader industry rather than Stability AI alone. The company remains valued at roughly $1 billion — unchanged from 2022 — but the combination of debt elimination, triple-digit revenue growth, and expanding product portfolio suggests a business model that is finally stabilizing after years of turbulence.

What Leaders Should Know

Stability AI offers the most comprehensive open-weight creative AI platform, spanning image, video, audio, 3D, and code generation. For organizations that need self-hosted creative AI — avoiding API dependencies, data sharing, or content restrictions — Stability AI's open-weight models remain the most mature option. Brand Studio provides a commercially supported path for enterprise creative production with brand governance. The Getty Images trademark ruling creates some legal uncertainty for commercial use of outputs, though the implications extend to the broader generative AI industry. Stability AI is best positioned as a component in a broader creative AI strategy, particularly for organizations that value open-source flexibility and on-premises deployment.