OpenAI CEO Sam Altman has described the film *Her*, in which Scarlett Johansson voices an empathetic AI assistant, as “prophetic.” However, Johansson publicly accused OpenAI of appropriating her voice for the company’s Sky voice assistant without her consent. Johansson, who had declined Altman’s offer to become the voice of ChatGPT, expressed that she was “shocked, angered, and in disbelief” upon hearing the assistant’s voice, which she said was indistinguishable from her own. Following the backlash, OpenAI discontinued the use of the voice. The incident highlighted concerns over how Silicon Valley technology firms handle the voices, likenesses, and rights of women.
The controversy is not isolated to OpenAI. Meta’s AI-powered Ray-Ban smart glasses faced criticism and earned the nickname “pervert glasses” amid fears of privacy violations and harassment. Similarly, law enforcement officials in the United States have been accused of misusing surveillance technologies such as Flock cameras to stalk former partners. Elon Musk’s SpaceXAI also received global condemnation when its chatbot Grok generated unsolicited and sexualized imagery of women and children, raising ethical questions about consent and AI misuse.
These instances have fueled broader debates about the accountability of corporations and developers in the creation and deployment of AI models, particularly their respect for women and ethical standards. Advocates argue that current AI systems reflect entrenched inequalities and power imbalances, motivating calls for fundamentally rethinking AI design and governance.
Some researchers propose a feminist approach to AI, emphasizing principles of connection, community, transparency, and equity. For example, the Data Against Feminicide project, co-led by one of the article’s authors, has collaborated with human rights activists in Latin America since 2019 to automate the identification of gender-related violence cases. The AI system scans tens of thousands of news articles daily to compile databases on feminicide, supporting advocacy efforts and family resources. Crucially, this tool was co-designed with activists who retained control over their data and decided which tasks were appropriate for automation, aligning with feminist values of consent and ownership.
In the United States, researchers like Lauren Klein utilize AI to highlight the contributions of women—especially Black women—to social justice movements, including abolition efforts. Their work employs open-source, transparent machine-learning models that avoid copyrighted materials, situating historical narratives within their relevant contexts. This approach underscores the idea that AI should complement human expertise rather than replace it, fostering a collaborative and informed use of technology.
Advocates argue that realizing a feminist AI ecosystem requires disrupting the dominance of Big Tech by developing public infrastructures governed by accountability and transparency instead of shareholder interests. Such a system would aim to support civic institutions and equitably distribute benefits and risks.
Grassroots efforts demonstrate public willingness to challenge current AI practices. In Quilicura, Chile, residents collectively responded to thousands of AI-generated prompts to draw attention to local environmental impacts of AI infrastructure. Meanwhile, Boston-area academics have engaged in symbolic acts, such as adorning surveillance cameras with decorative elements, to spark dialogue about AI surveillance.
These cases suggest that pressing for a more ethical and equitable AI future is possible, with advocates encouraging sustained activism to transform the digital landscape. As political activist Angela Davis once stated, change requires persistent belief in and work toward radical transformation—a principle that underpins calls for feminist AI development.