The Georgetown-IBM experiment, held at the IBM head office in New York City in the United States, offers the first public demonstration of machine translation. The system itself, however, is no more than what today would be called a "toy" system, having just 250 words and translating just 49 carefully selected Russian sentences into English — mainly in the field of chemistry. Nevertheless, it encourages the view that machine translation was imminent — and in particular stimulates the financing of the research, not just in the US but worldwide.
In 1958, linguist Yehoshua Bar-Hillel travels around the world visiting machine translation centers to better understand the work they were doing. In 1959, he writes up a report (intended primarily for the US government) pointing out some key difficulties with machine translation that he believed might doom the efforts then underway. An expanded version of the report is published in 1960 in the annual review journal Advances in Computers. His main argument was that existing methods offered no way of resolving semantic ambiguities whose resolution required having an understanding of the terms being used, such as the ambiguity arising from a single word having multiple meanings.
ALPAC publishes a report commissioned by the United States government. The report concludes that machine translation is more expensive, less accurate and slower than human translation, and that despite the expenses, machine translation is not likely to reach the quality of a human translator in the near future. It recommends that tools be developed to aid translators — automatic dictionaries, for example — and that some research in computational linguistics should continue to be supported. The report causes a significant decline in government funding for machine translation in the US, and to a lesser extent in the UK and Russia.
The METEO System, developed at the Université de Montréal, is installed in Canada to translate weather forecasts from English to French, and is translating close to 80,000 words per day or 30 million words per year until it is replaced by a competitor's system on 30 September 2001.
Makoto Nagao proposes example-based machine translation. The idea is to break down sentences into phrases (subsentential units) and learn the translations of those phrases using a corpus of examples. With enough phrases known, new sentences that combine existing phrases in a novel manner can be translated.