The Unseen AI: Beyond Sci-Fi Scares
While headlines often fixate on hypothetical superintelligence threats embodied by fictional robots like Terminator, a more immediate concern is emerging: the AI we already use. Samuel Segun, a Nigerian philosopher and AI safety researcher, argues that our focus should shift from speculative futures to the tangible impact of AI systems currently integrated into our lives. Segun's career, spanning robotics in his youth to advising the UN on AI, is driven by a singular question: how should AI function effectively and ethically in the real world?
This perspective challenges the dominant narrative that often frames AI risks as distant, existential threats. Instead, Segun points to the present, suggesting that the AI we interact with daily—in recommendation engines, content generation tools, and decision-making algorithms—carries its own set of ethical and practical challenges. The danger isn't necessarily a rogue AI uprising, but the subtle, pervasive influence of AI systems that shape our perceptions, influence our choices, and automate our tasks, often without our full understanding or consent.
Segun's philosophical grounding provides a crucial lens for examining these systems. Rather than viewing AI as a purely technical problem, he approaches it as a multifaceted issue with deep societal implications. This includes questions of bias, fairness, accountability, and the very definition of intelligence and creativity in an increasingly automated world. His work emphasizes that understanding and guiding the development of AI requires not just engineering prowess, but also a robust ethical and philosophical framework.
Creation vs. Transformation: A New AI Paradigm
A key insight emerging from discussions around practical AI use, highlighted by Segun's philosophical approach and user experiences, is the concept of creation < transformation. This paradigm shift suggests that the most effective and perhaps most ethical way to leverage AI is not by asking it to generate novel content from scratch, but by tasking it with transforming existing information or ideas.
Consider the process. Instead of prompting an AI to write a complete article, a developer might provide rough bullet points and ask the AI to flesh them out into prose. A writer might feed existing text into an AI and request a summary, a change in tone, or adaptation for a different audience. Programmers can present existing code and ask for modifications to suit a new environment. This approach treats AI less like an oracle of original thought and more like an exceptionally capable assistant for refinement, adaptation, and reformatting.
This method offers several advantages. Firstly, it grounds AI's output in human intent and existing context, reducing the likelihood of generating nonsensical or factually inaccurate content that often plagues purely generative prompts. Secondly, it respects the user's role as the primary source of creativity and direction. The AI becomes a tool for augmenting human capabilities, not replacing them entirely. This collaborative model fosters a more productive and less fraught relationship with AI technologies.
Navigating the Real-World AI Landscape
Segun's advocacy for AI safety extends beyond theoretical discussions to practical advice for organizations and policymakers. His work with the UN and his experience building industrial AI systems provide him with a unique perspective on the challenges of deploying AI in sensitive or critical environments. This involves not only ensuring the technical robustness of AI models but also understanding their potential downstream effects on employment, social structures, and individual autonomy.
The focus on transformation over pure creation aligns with this practical outlook. Industrial AI often involves optimizing existing processes, analyzing large datasets for insights, or automating repetitive tasks within established workflows. These applications are less about generating entirely new outputs and more about improving the efficiency, accuracy, and scope of human endeavors. By framing AI as a transformative tool, we can better manage its integration into complex systems, ensuring that it serves as an enabler rather than a disruptive force.
What nobody has addressed yet is the long-term impact on human skill development when AI predominantly functions as a transformer. If we increasingly rely on AI to refine our prose, adapt our code, or reframe our ideas, do we risk atrophying our own abilities in these areas? The convenience of AI-driven transformation could inadvertently lead to a deskilling effect, making us more dependent on the technology and less capable of performing these tasks independently.
Ethical Considerations in an AI-Saturated World
The philosophical underpinnings of Segun's work are crucial for navigating the ethical minefield of current AI applications. When AI systems are used for transformation, they are often working with data that reflects existing societal biases. An AI tasked with summarizing news articles, for instance, might inadvertently perpetuate biases present in the original reporting if not carefully guided. Similarly, an AI transforming code might introduce subtle security vulnerabilities if its training data or algorithms are flawed.
Segun's emphasis on real-world AI safety means confronting these issues head-on. It requires transparency in how AI models are trained, rigorous testing for bias, and clear lines of accountability when AI systems produce harmful outcomes. The