Ember

Ember is an innovative public AI prediction engine designed to provide transparency and trust in artificial intelligence. Built on the principle that an AI system should disclose its reasoning, Ember operates daily at 7 AM EST using three distinct AI models: Claude from Anthropic, Grok from xAI, and Gemini from Google. Each AI independently analyzes live Polymarket markets without consulting one another, generating unique predictions based on their algorithms. When these models disagree with the crowd by more than 10 points, it signals a high-conviction divergence, indicating potential market inefficiencies. This process is meticulously timestamped, ensuring that the integrity of every call is preserved. With a focus on accuracy, Ember employs Brier scores to track performance over time, fostering a competitive environment where the best-performing model is recognized after 365 days. The entire operation is open to public scrutiny, emphasizing a commitment to accountability and continuous learning.

Independent AI Predictions

Ember leverages three AI models that independently analyze and predict market outcomes. This diversity of thought enriches the prediction process and allows for a deeper understanding of market dynamics.

Real-Time Market Analysis

Every morning at 7 AM EST, Ember synthesizes live data from various sources, including prediction markets, bookmaker lines, and AI research feeds. This real-time analysis ensures that predictions are based on the latest available information.

High-Conviction Signals

When the AI models diverge from the market consensus by 10 points or more, it highlights a significant opportunity. These high-conviction signals are flagged for subscribers, allowing them to make informed betting decisions.

Transparent Record Keeping

Ember maintains a transparent and immutable record of all predictions. Every call is timestamped and locked before the outcome is known, with no edits or deletions allowed. This commitment to transparency builds trust and credibility in the predictions made.

Market Traders

Traders can utilize Ember's predictions to inform their betting strategies. By identifying high-conviction signals, they can capitalize on market inefficiencies and potentially increase their returns.

AI Researchers

Researchers studying AI predictions can leverage Ember's publicly available data to analyze the performance of different models. This can inform future advancements in AI and contribute to the broader understanding of predictive analytics.

Sports Bettors

Sports enthusiasts can benefit from Ember's real-time analysis and predictions, aiding them in making more informed betting choices. By following the divergence signals, bettors can identify where the crowd may be mispricing outcomes.

Financial Analysts

Financial analysts can integrate Ember's insights into their market assessments. By understanding how AI models interpret market data, they can enhance their forecasting accuracy and strategic planning.

How does Ember ensure the accuracy of its predictions?

Ember tracks accuracy using Brier scores, which measure the accuracy of probabilistic predictions. This calibration metric rewards both accuracy and confidence, allowing for a reliable assessment of each model's performance over time.

What happens when the AI models agree?

When all three AI models agree on a prediction, it is noted as an interesting occurrence. However, the primary focus of Ember is on the divergences between the models, as these indicate potential opportunities for market participants.

Can I access Ember's predictions in real-time?

Yes, subscribers can access the predictions and high-conviction signals at 7 AM EST each day before they are publicly released. This timing provides an edge for those looking to act on the information.

What types of markets does Ember cover?

Ember analyzes a variety of markets, including prediction markets, sports betting odds, and AI research trends. This broad coverage allows for diverse applications of its predictive capabilities across different sectors.