AI Censorship Fear, Is Our Knowledge Being Controlled?
- Charles Hoskinson highlights AI censorship eroding technology's utility.
- AI models show differing responses to potentially dangerous information.
- Elon Musk also expresses his concerns about rapid AI development and its risks.
Charles Hoskinson on AI Censorship Eroding Utility and Innovation
Charles Hoskinson, the founder of Cardano (ADA) has raised significant concerns about the increasing trend of Artificial Intelligence (AI) censorship. He believes that the current approach to AI censorship is diminishing the technology's utility and altering societal perspectives. This development has sparked a broader conversation among tech leaders and innovators about the future of AI.
What's the News
In a recent post on X, Charles Hoskinson highlighted a troubling trend in AI technology censorship. He emphasized that AI censorship, particularly through alignment training is gradually eroding the technology's usefulness. Hoskinson pointed out that a small, unelected group is deciding what knowledge is permissible for the public, comparing it to the forbidden knowledge imposed on future generations.
To illustrate his point, Hoskinson shared screenshots of two prominent AI models responding to a specific query about constructing a Farnsworth fusor. ChatGPT 4.0, a leading AI model acknowledged the device's potential danger and the need for expert supervision. Despite this, it provided a list of components required to build the fusor. Similarly, Anthropic’s Claude 3.5 Sonnet offered general information about the device, its components, and its history, while cautioning about its potential hazards.
These responses highlight apprehensions of Hoskinson regarding the issue of AI censorship, mirroring the wider concerns expressed by other esteemed thought leaders and technology experts.
Lately, there has been a notable emergence of concerns from both current and former employees of organizations such as OpenAI, Google DeepMind, and Anthropic. These concerns include the rise of misinformation, the diminishing control over autonomous AI systems, and the potential threat of human extinction.
Despite these concerns, the development and introduction of new AI tools continue. Robinhood recently launched Harmonic, a commercial AI research lab focused on solutions related to Mathematical Superintelligence (MSI) further emphasizing the ongoing advancements in AI.
Elon Musk's Words on AI
Elon Musk believes that AI is shifting societal perspectives by prioritizing political correctness over truth. He argues that this bias in AI systems can lead to dangerous outcomes, as seen with Google’s Gemini AI. Musk warns that training AI to lie for the sake of political correctness could have disastrous consequences. To counter this, he founded xAI, aiming to develop AI that is truthful and inquisitive. Musk asserts that only AI committed to truth, no matter how uncomfortable, can truly benefit humanity and foster genuine progress.
Conclusion
Many Experts including Charles Hoskinson’s concerned about AI censorship. The hot topic in this context is a balance of safety and innovation in AI as more voices call for careful consideration of the technology’s potential risks and benefits. Elon Musk also stated his views recently on Politically correct AI rather than being true. This will also create a problem in the future because AI is shifting the perspective of society. As AI continues to evolve, discussions surrounding its ethical application and regulation are expected to grow, shaping the future of this transformative technology.
