Making foundation models accessible: The battle between closed and open source AI
The rise of foundation models has fundamentally shifted how humans interact with data, raising critical questions about privacy, consent, and control. As these powerful systems redefine technological interfaces, the debate over who governs them becomes paramount. The concentration of this power in a few large, closed-source corporations poses significant ethical risks, making the accessibility of AI infrastructure a central concern for society. This tension highlights the urgent need for open-source alternatives to democratize access and prevent monopolistic control. While initiatives like Stable Diffusion have inspired developer communities by placing technology in the public’s hands, sustaining open-source foundation models remains financially daunting. High computational costs and the lack of proven long-term profitability deter venture capital, often forcing open projects to rely on Big Tech investment, which contradicts the goal of independent development. For the open_data community, this landscape defines the struggle between ethical transparency and economic viability. The article underscores that open-source AI must evolve into sustainable businesses to avoid being overshadowed by well-funded proprietary giants. Success requires not just technical innovation, but robust governance models that address ethical responsibilities. Ultimately, ensuring open, accountable AI systems survive is essential to preserving public trust and equitable access to transformative technology.
Source: techcrunch.comPublished on 2023-05-12
Related news
- Cédula de privacidad
- AI2 is developing a large language model optimized for science
- Las comunicaciones anónimas y su régimen en la Ley de protección del informante (“whistleblower”)
- We've received 1,053 RTI requests, 838 approved - Ghanamma.com
- Pobreza subió: Uno de cada tres peruanos no puede costear una canasta básica, según el INEI | RPP Noticias