Hybrid Intelligence - Combining human intuition and AI to power next-generation decisions

Hybrid Intelligence - Combining human intuition and AI to power next-generation decisions

Hybrid Intelligence? is a new class of technology built to solve the world's hardest problems by combining the complementary strengths of humans and AI (Dellermann et al. 2019). Powered by human intuition. Augmented by AI. Hybrid Intelligence? can overcome the limitations of existing AI technology in complex business contexts and augment decisions at scale.

How does it work?

Hybrid Intelligence? combines human intuition with the power of AI to empower outperforming decision-making. The goal is not to replace humans but to lift their capabilities to the next level. It is defined along three dimensions:

Collective problem-solving

Human experts and AI collaborate during business workflows and use their complementary abilities to augment each other.

Superior performance

Combining humans and AI empowers us to solve complex problems that could not be solved before at a higher performance.

Continuous learning

Both humans and machine learning models continuously learn from each other and improve over time.

What problem does it solve?

Today's business world requires a new kind of technology to keep up with the context in which decisions are made. Thus, despite the enormous progress in AI technology, it could not be applied in many business settings. There are four reasons why existing AI technology alone is not sufficient for this domain.

Uncertainty

The decision context is too uncertain to predict the future solely from historical data and includes many unknown unknowns or "black swan" events. Examples include innovation or strategy decisions that require imagining a new future.

Complexity

The decision context is too complex and the data that you need to train machine learning models are too ill-structured to be used by AI alone. Examples include the failure of AI models in highly connected business decisions.

Fast-changing

The decision context is too dynamic, and constant re-training of machine learning models is neither feasible nor profitable. Examples include the failure of sales forecasting methods after the outbreak of COVID-19.

Cost of error

If the stakes of the decision context are high or regulated and the cost of errors is enormous and full automation is way too dangerous. Examples include investment or acquisition decisions or other strategic actions.

Learn more about the latest in Hybrid Intelligence? research

要查看或添加评论,请登录

社区洞察

其他会员也浏览了