The Proto-Altaic language is a hypothetical extinct language that has been proposed as the common ancestor of the disputed Altaic languages.

In the 18th century, some similarities between the Turkic, Mongolian, and Tungusic languages led to the conjecture that they would be a single language family with a common ancestral language.^{: page 125} Starting in the 19th century, some linguists proposed to include also the Japonic and/or Koreanic languages as well as the Ainu language, forming what would later be called the "Macro-Altaic family" (the original one being then dubbed "Micro-Altaic").^: 34 Around the same time others proposed to include the Uralic languages in a Ural-Altaic family.^{: 126–127}

Versions of the Altaic family hypothesis were widely accepted until the 1960s, and it is still listed in many encyclopedias and handbooks.^: 73 However, in recent decades, the proposal has received substantial criticisms and has been rejected by many comparative linguists.

Nevertheless, "Altaicists" (supporters of the theory of a common origin for the Altaic languages) such as Václav Blažek and Sergei Starostin have endeavored to reconstruct "Proto-Altaic," the hypothetical common ancestral language of the family.

Some Altaicists have proposed that the original area where Proto-(Macro-)Altaic would have been spoken was a relatively small area comprising present-day North Korea, Southern Manchuria, and Southeastern Mongolia. The date for its split into the major recognized families was estimated at around 5,000 BC or 6,000 BC. This would make Altaic a language family about as old as Indo-European (4,000 to 7,000 BC according to several hypotheses) but considerably younger than Afroasiatic (c. 10,000 BC^: 33 or 11,000 to 16,000 BC^: 35–36 according to different sources).