International Joint Conferences on Artificial Intelligence (IJCAI)

What IJCAI is

IJCAI — the International Joint Conferences on Artificial Intelligence — is a non-profit corporation founded in California in 1969 for scientific and educational purposes^[1]. The abbreviation is used both for the organisation and for the conference series it runs. Being a non-profit corporation registered in California means the body has no owners drawing profit out of it, and its charter commits it to stated scientific and educational goals rather than commercial ones.

Stated purposes and how it is organised

The organisation’s stated purposes include disseminating information on artificial intelligence at conferences and through proceedings, books, video recordings and other educational materials^[1]. Internally it consists of two divisions, the Conference Division and the AI Journal Division^[1]: one looks after the meetings, the other after the journal. It is governed by a Board of Trustees, with an IJCAI Secretariat in charge of its operations^[1]. A governing board plus a small permanent secretariat is a common shape for learned societies — the board sets policy, the secretariat carries out the day-to-day work.

The conference series

IJCAI conferences were held biennially in odd-numbered years from 1969, and annually since 2015^[1]. Moving from a two-year to a one-year cycle roughly doubles how often the community meets under this name. Locations and dates of future conferences are announced officially at the preceding conference^[1], so that the next venue becomes known to the people attending the current one — a practical arrangement for a meeting large enough that people plan attendance well ahead.

The proceedings as a record

The organisation’s proceedings archive lists every conference from the first, in Washington in 1969, to the most recent^[2]. The first proceedings were edited by Donald E. Walker and Lewis M. Norton, and cover sessions on man-machine symbiosis, pattern recognition, heuristic problem solving, theorem proving and integrated AI systems^[2]. Man-machine symbiosis was the phrase of the period for people and computers working closely together; heuristic problem solving and theorem proving name two of the problems that early artificial intelligence research was built around, and integrated AI systems refers to attempts to combine several such capabilities inside one program.

After that first volume the archive records the second conference in London in 1971, the third at Stanford in 1973, the fourth in Tbilisi in 1975 and the fifth in Cambridge in 1977^[2]. A proceedings volume is more than an administrative record: it is the part of a conference that survives afterwards, the place where a talk given in a lecture room becomes a paper that later researchers can read and cite. Keeping an archive that begins with the first meeting makes that record continuous.

Other material

The organisation’s site also points to AI Hub and to free access to the AI journal^[1]. In practice this means the organisation distributes material in several forms at once — a meeting, a printed and archived proceedings series, a journal, and separate channels for reaching people who did not attend the conference.