AI and computer chess history: in 1985, International Master David Levy defeated a reigning world-champion chess program in four straight games by adopting orthodox play that confused the engine. What was the name of that program?

Introduction / Context:
Computer chess has long served as a benchmark for artificial intelligence. In the mid-1980s, programs achieved master-level performance but still showed weaknesses that strong human players could exploit. Remembering landmark matches helps situate the evolution from early rule-based and alpha–beta programs to modern engines powered by advanced evaluation and, later, machine learning.

Given Data / Assumptions:

• The match occurred in 1985 and featured David Levy, a noted computer-chess critic and International Master.
• The program was a world-champion engine of its time.
• The human strategy relied on orthodox, non-speculative play to induce programmatic misjudgments.

Concept / Approach:

During this period, CRAY BLITZ was among the strongest programs, winning the World Computer Chess Championship (WCCC) and running on Cray supercomputers. Levy's strategy highlighted evaluation and horizon effect limitations common to engines of the era. While Kaissa was an earlier Soviet champion (1974), and other names listed are not world-champion chess programs, CRAY BLITZ aligns with the timeline and status indicated in the stem.

Step-by-Step Solution:

Verification / Alternative check:

Contemporary articles and retrospectives discuss Levy's victories over top programs of that epoch, including CRAY BLITZ, prior to the dominance of later systems like Deep Thought and Deep Blue.

Why Other Options Are Wrong:

Kaissa: Earlier champion, not matched to the 1985 event described.

Golf/DIGDUG: Not computer-chess world champions.

None: Incorrect because CRAY BLITZ fits the facts.

Common Pitfalls:

Confusing different championship years or mixing later engines with earlier-era matches.

Final Answer:

CRAY BLITZ