AI winter

from Wikipedia, the free encyclopedia

In the history of artificial intelligence , phases are referred to as AI winters in which public and private research funding , start-up financing and investments in artificial intelligence (AI) decline sharply. This goes hand in hand with little progress on the part of the subject, at least compared to the (exaggerated) expectations aroused previously. In this respect, the AI ​​winter is an example of the “valley of disappointments” in the hype cycle . A high level of media attention in combination with the often vague and poorly selective definitions of artificial intelligence can be seen as a potential cause of the excessive expectations that are often circulating about AI technologies.

The word creation AI winter comes from the American . The term is partly interpreted as an analogy to the nuclear winter and partly attributed to the Lisp developer Richard P. Gabriel . The other interpretation of the term refers to the season - the cold comes, there is no more growth, but after winter comes spring again.

The phase from the beginning of the 1970s is known as the first AI winter. One trigger was a book by Marvin Minsky and Seymour Papert on perceptrons from 1969 that revealed the limits of this approach. An evaluation carried out by James Lighthill on behalf of the British Parliament in 1973 caused even greater disillusionment ( Lighthill report ). As a result, DARPA , among others, cut its funding for AI significantly.

After the field recovered in the 1980s, high hopes were placed in expert systems and Lisp machines . The second AI winter began in 1987 with the disappointment of high expectations for this approach. This development was predicted in 1984 by Roger Schank at the annual meeting of the American Association of Artificial Intelligence (AAAI).

Web links

Individual evidence

  1. ↑ Retresco: What is an AI winter? In: Retresco. September 26, 2019, accessed on October 9, 2019 (German).
  2. ↑ Paul E. Ceruzzi: Manned Spaceflight and Artificial Intelligence: "Natural" Trajectories of Technology . In: David L. Ferro, Eric G. Swedin (Eds.): Science Fiction and Computing: Essays on Interlinked Domains . McFarland, Jefferson (NC) 2011, ISBN 9780786489336 , p. 106.
  3. ↑ John Paul Mueller, Luca Massaron: Artificial Intelligence For Dummies . John Wiley & Sons, Hoboken (NJ) 2018, ISBN 9781119467625 , pp. 230f.
  4. ^ M. Tim Jones: Artificial Intelligence: A Systems Approach . Jones & Bartlett, Sudbury (MA) 2015, ISBN 9781449631154 , p. 8.
  5. ↑ James Lighthill (1973): "Artificial Intelligence: A General Survey" in Artificial Intelligence: a paper symposium, Science Research Council
  6. ↑ Amy J. Connolly, T. Grandon Gill (Eds.): Debates in Information Technology . Muma Business Press, Santa Rosa (CA) 2015, ISBN 9781681100005 , pp. 108-109.
  7. ↑ Amy J. Connolly, T. Grandon Gill (Eds.): Debates in Information Technology . Muma Business Press, Santa Rosa (CA) 2015, ISBN 9781681100005 , pp. 110-111.
  8. ^ The dark ages of AI: A panel discussion at AAAI-84 . In: AI Magazine . Volume 6 Issue 3 (January 1985), pp. 122-134. ( online )