Communications of the ACM
There Was No 'First AI Winter'
Credit: Bravissimos
As I concluded my June Historical Reflections column, artificial intelligence had matured from an intellectual brand invented to win funding for a summer research workshop to one of the most prestigious fields in the emerging discipline of computer science. Four of the first 10 ACM A.M. Turing Award recipients were AI specialists: Marvin Minsky, Herb Simon, Allen Newell, and John McCarthy. These men founded the three leading AI labs and played central roles in building what are still the top three U.S. computer science programs at MIT, Stanford, and Carnegie Mellon. Conceptually AI was about uncovering and duplicating the processes behind human cognition; practically it was about figuring out how to program tasks that people could do but computers could not. Although connectionist approaches based on training networks of simulated neurons had been prominent in the primordial stew of cybernetics and automata research from which AI emerged, all four Turing Award recipients favored the rival symbolic approach, in which computers algorithmically manipulated symbols according to coded rules of logic.
A History of Failed Ideas?
AI was born in hype, and its story is usually told as a series of cycles of fervent enthusiasm followed by bitter disappointment. Michael Wooldridge, himself an eminent AI researcher, began his recent introduction to the field by remembering when he told a colleague about his plan to tell "the story of AI through failed ideas." In response, "she looked back at me, her smile now faded. 'It's going to be a bloody long book then.'"22
Comments
Matthias Felleisen
Is it necessary to characterize the US and all research efforts in the US as "imperialistic" and "capitalistic"? If we wrote a history of computing research in the Soviet Union would we add "colonialist" (as in "third world") and "communist" (as in evil) to the history? We should count ourselves lucky that all this research made a (small) contribution to ending the Cold War.
Herbert Bruderer
Five years before the famous 1956 Dartmouth meeting, a large, well-documented European conference on non-numerical data processing was held in Paris, see
The Birthplace of Artificial Intelligence? | blog@CACM | Communications of the ACM
Read more in:
Bruderer, Herbert: Meilensteine der Rechentechnik, De Gruyter Oldenbourg, Berlin/Boston, 3. Auflage 2020, Band 1, 970 Seiten, 577 Abbildungen, 114 Tabellen, https://doi.org/10.1515/9783110669664
Bruderer, Herbert: Meilensteine der Rechentechnik, De Gruyter Oldenbourg, Berlin/Boston, 3. Auflage 2020, Band 2, 1055 Seiten, 138 Abbildungen, 37 Tabellen, https://doi.org/10.1515/9783110669671
Bruderer, Herbert: Milestones in Analog and Digital Computing, Springer Nature Switzerland AG, Cham, 3rd edition 2020, 2 volumes, 2113 pages, 715 illustrations, 151 tables, translated from the German by John McMinn, https://doi.org/10.1007/978-3-030-40974-6
Displaying all 2 comments