Why you need a shaman to design and work on AI projects?

The notion of needing a shaman to design and work on AI projects is not rooted in professional AI development practices or the science behind artificial intelligence. Shamans are typically understood as spiritual guides in various indigenous cultures, not as experts in technology or computer science.

However, if we look at this idea metaphorically or symbolically, one might argue that a "shaman" could represent someone who brings a unique perspective to an AI project, perhaps one that is deeply intuitive, holistic, or connected to human experience in a way that typical scientific approaches are not. Here are a few abstract qualities a "shaman-like" figure might metaphorically bring to an AI project:

1. Holistic Perspective: Shamans traditionally take a holistic approach to their practice, integrating the well-being of the individual with the larger community and environment. Similarly, in AI, a team member with a holistic perspective might ensure that the AI systems are designed with consideration for their broader impact on society and the environment.
2. Ethical Guidance: Shamans often serve as moral and spiritual guides. In AI development, having someone focused on ethics can help guide the project to ensure that the AI systems are developed responsibly and do not perpetuate bias, inequality, or harm.
3. Intuition: Shamanic practice often involves a degree of intuition. An intuitive approach to problem-solving in AI could help to identify solutions that are not immediately apparent through logical analysis alone.
4. Interdisciplinary Insight: Shamans usually have knowledge that spans across different areas of their culture, including medicine, storytelling, and conflict resolution. Similarly, an AI project can benefit from team members with interdisciplinary knowledge, bringing insights from fields like psychology, philosophy, or the arts to create more innovative and human-centered technologies.
5. Healing/Therapeutic Perspective: Shamans are healers. In a metaphorical sense, someone who considers the potential of AI to heal or enhance human life could push projects in a direction that emphasizes care, well-being, and support.
6. Connection to Tradition and Storytelling: Shamans are also storytellers who preserve and share cultural narratives. An AI "shaman" could ensure that the technology respects and incorporates the stories, languages, and traditions of diverse cultures, which is important in creating inclusive and culturally aware AI systems.

While the above qualities could be valuable, it is crucial to note that actual AI development is highly technical and requires specialized knowledge in computer science, mathematics, data analysis, machine learning, and domain-specific expertise. Therefore, while individuals with diverse backgrounds and perspectives can contribute significantly to the richness and ethical grounding of AI projects, the core design and development work will still rely on trained AI professionals.

What rituals will a Shaman perform to create AI systems?

We can imagine a set of "rituals" or practices that symbolically resemble the role of a shaman in ensuring that the AI system is developed with a sense of purpose, ethics, and consideration for its impact:

1. Setting Intentions: Before starting the development process, the team might gather to set clear intentions for the AI system's role and the values it should embody. This is akin to setting the vision or the spirit of the project.
2. Blessing the Data: Data is the lifeblood of AI. A "ritual" might involve thoroughly reviewing and "cleansing" the data to remove biases, ensure privacy, and respect the sources of the data, much like a shaman might cleanse a space of negative energies.
3. Invoking Multiple Perspectives: Just as a shaman might invoke the wisdom of ancestors or the community, the development team could ensure that multiple stakeholder perspectives are considered in the design and implementation of the AI system.
4. Ethical Guidance Ceremony: The team could hold regular meetings where the ethical implications of the AI system are discussed, similar to how a shaman might guide their community through moral dilemmas.
5. Respecting the Elements: In shamanistic practices, there's often a respect for natural elements—earth, air, fire, water. In AI, this could translate to a respect for the elements of the system: hardware (earth), data flow (air), computing power (fire), and cooling systems (water), ensuring that each is sourced and used sustainably.
6. Journeying for Insights: Shamans often undertake journeys to gain insights. In AI, this could be metaphorically applied to exploring different models and algorithms, diving deep into the 'spiritual' landscape of machine learning to find the one that fits the project's needs the best.
7. Integration Ritual: Integrating the AI system into its intended environment could be seen as a ritual, ensuring that it aligns with human workflows, complements human abilities, and respects human dignity.
8. Continuous Reflection and Adaptation: Much like shamans who continually reflect on their experiences and adapt their practices, AI developers need to monitor, update, and maintain AI systems over time to ensure they continue to operate ethically and effectively.

While these metaphorical "rituals" are not part of actual AI development practices, they do illustrate the importance of intentionality, ethics, and the broader societal and environmental impact in the AI field. The real-world equivalent to ensuring responsible AI development involves interdisciplinary collaboration, ethical guidelines, regulatory compliance, and ongoing engagement with stakeholders.

Two stories

Story 1: The Data Healer

Company Challenge: A healthcare tech company was struggling with its AI diagnostic tool. The AI was failing to correctly diagnose a range of conditions, and the company couldn’t figure out why. They had data scientists and medical experts on the team, but the integration of diverse datasets was problematic.

The Shaman's Ritual: The company sought the help of a "Data Healer," a shaman known for her ability to "communicate" with data. She began her ritual with a gathering of all the project stakeholders. Sitting in a circle, she asked each person to express their intention for the AI system, grounding the project with a unified vision of healing and support for patients.

The Data Healer then "journeyed" into the depths of the datasets. This journey was a deep dive analysis where she examined the data sources, structures, and relationships. She looked for patterns and connections that were not immediately obvious, seeking insights from the "ancestors" of the data – its origins, collection methods, and the cultural contexts from which it was derived.

Outcome: Through her ritual, the Data Healer uncovered that the AI’s diagnostic issues were due to cultural biases present in the training data. She guided the data scientists to balance the datasets with more diverse data, which included a wider range of symptomatic expressions across different cultures. With her insights, the AI diagnostic tool became more accurate and culturally sensitive, leading to better patient outcomes.

Story 2: The Algorithm Whisperer

Company Challenge: A financial services firm wanted to use AI for predicting market trends but found that its models were overly volatile and often inaccurate. They needed a new perspective to understand the unpredictable nature of financial markets.

The Shaman's Ritual: The firm engaged the services of an "Algorithm Whisperer," a shaman with the reputed ability to "tune in" to the algorithmic spirits. The Whisperer started his ritual by meditating with the AI development team, focusing on the intention of creating an AI that would bring prosperity and stability.

He then embarked on a virtual "vision quest" within the machine learning environment. The Whisperer immersed himself in the company’s historical data and algorithms, looking for hidden messages and signs in the ebb and flow of market data. He used visualization techniques and pattern recognition to identify the underlying structures of market movements that the AI algorithms could learn from.

Outcome: The Algorithm Whisperer revealed that the AI models were too rigid and needed to adapt more fluidly to market changes. He recommended the implementation of ensemble methods that could integrate multiple predictive models, much like a council of elders, each bringing its wisdom to the decision-making process. The firm adopted this approach, and the resulting AI system provided more stable and reliable market predictions, enhancing the firm’s ability to advise its clients.