Claude, Anthropic's AI, Shines in Creative and Unexpected Roles

Claude, Anthropic's AI, and Its Surprising Responses

Every day, I turn to Claude, the AI assistant developed by Anthropic, with a multitude of questions. While most of the answers are helpful, certain specific prompts trigger particularly brilliant responses. By applying a few simple tricks, I discovered that Claude could transform a mundane question into an astonishingly insightful explanation.

Claude stands out in the world of AI assistants due to its ability to provide nuanced and well-structured answers. This reputation largely rests on an innovative approach called "constitutional AI." This unique system allows Claude to examine its own responses through guiding principles before formulating them.

However, my daily experience with Claude has revealed a crucial aspect: the quality of the answers is often directly linked to how the question is posed.

When the question is asked vaguely, Claude's response is generally correct but lacks originality. In contrast, by assigning it a specific role or creative framework, Claude becomes much more interesting. Here are five methods I use almost daily to elicit fascinating responses.

The Pizza and Pirates Method to Demystify Blockchain

Claude particularly excels when creative constraints are imposed. This pushes it to develop original metaphors to explain complex topics.

Take the example of blockchain. Instead of asking for a classic technical explanation, I opted for a playful approach: "Explain how blockchain works using only metaphors involving pizza delivery and pirates."

Claude then imagined a captivating educational scenario. It described a fleet of pizza delivery drivers crisscrossing the city, each noting every order in their notebooks. Each driver has an identical copy of the information, and if one of them tries to cheat, the others immediately detect the alteration.

In this metaphor, the pirates come into play by comparing each page of the ledger to a treasure book sealed by a pirate captain. Each page depends on the seal of the previous one, making the concept of a distributed ledger accessible even to those without technical knowledge.

Simplifying Complex Topics with "The Retired Professor"

To understand a complex topic, I often ask Claude to play a specific role. This technique completely alters how the explanation is constructed.

For example, I asked Claude: "Explain how inflation works as if you were a retired economics professor talking to a curious teenager."

Claude then told a story about a neighborhood bakery. The baker observes that more and more customers are lining up each morning, which leads to a gradual increase in the price of bread. Nothing changes in the recipe, but the growing demand is the cause.

This imagery makes inflation much more tangible, transforming it from an abstract economic theory into a daily phenomenon.

The Future Historian Technique to Narrate an Era

I also enjoy using narrative roles to obtain more creative responses. Claude excels in this type of scenario.

I proposed a simple exercise: "Imagine you are a historian in the year 2100 explaining the rise of social media."

Claude's response resembled an anthropology lecture. It described the platforms as rival digital cities. "For students in the 22nd century, social media resembles competing digital kingdoms vying for attention," it explained.

Influencers have become a merchant class, while algorithms appear as invisible bureaucrats. This perspective transforms a technological explanation into a genuine historical narrative.

Organizing an Impossible Debate Between Great Figures

Claude is also very effective when it comes to managing discussions between different viewpoints. I sometimes use this ability to simulate intellectual debates.

One day, I asked it: "Write a conversation between Albert Einstein and Isaac Asimov discussing artificial intelligence in a bar."

Claude then wrote a scene worthy of a small play. Einstein, cautious, reflects on the limits of computation. "A machine can calculate very quickly, but calculation does not necessarily create wisdom," he asserts. Asimov responds enthusiastically, imagining machines capable of exploring thousands of scientific hypotheses.

The result resembles a philosophical discussion rather than a simple AI response.

Requesting a Structured Analysis to Understand a Strategy

Claude also becomes very effective when asked for an organized analysis. For example, I asked it to explain the downfall of Blockbuster in the face of Netflix.

The prompt requested a detailed list of strategic errors. Claude immediately structured its response like a presentation for business leaders. It explained that Netflix viewed technology as the core of its business, while Blockbuster saw digital as merely a secondary project.

This difference in approach gradually widened the gap between the two companies.

After several months of use, I understood a very simple principle. Claude always responds better when given a clear role and a specific objective. A vague question often produces a mundane answer. However, a creative prompt immediately transforms the outcome.

Assign a role, a context, and a mission. Claude then becomes much more surprising and sometimes even brilliant.