In the rapidly evolving field of Artificial Intelligence, a growing debate centers around the concept of open-source AI and the challenges posed by selective transparency. On one hand, the idea of open-source AI suggests that all elements of an AI system should be accessible to the public for examination, experimentation, and understanding. However, in practice, this is rarely the case. At Weebseat, we recognize that what many regard as open-source often fails to provide complete transparency. It’s common to find that key components remain inaccessible, leaving developers and practitioners unable to fully understand the algorithms and data used in training these systems. Such opacity not only hampers innovation but also poses significant risks. Without full access, it’s challenging to assess the ethical implications, biases, or potential failings within AI systems. Our team believes that true open-source AI requires every element, from datasets to code, to be fully transparent. Selective transparency encourages a breed of AI development where proprietary constraints dictate the operational understanding of these powerful tools. The risks are manifold: unverified claims about AI capabilities, potential misuse, and unchecked biases that could perpetuate or exacerbate societal issues. Experts suggest that stakeholders in AI development need to advocate for frameworks that ensure complete transparency. Fostering a culture where every step of AI creation is documented, reviewed, and open to scrutiny will facilitate a more ethical and responsible development landscape. Until such an approach is widely adopted, the label of open-source AI remains misleading and problematic. Developing guidelines and regulations may encourage a shift towards genuine transparency, benefiting creators, users, and society at large.