The Creation Science Institute tells us in
How Do You Date a Volcano? by Robert Doolan:
Researchers found similar conflict in Hawaii. A lava flow that is known to have taken place in 1800-1801 less than 200 years ago was dated by potassium-argon as being 2,960 million years old. (3) If the real dates had not been reasonably well established by other means, who could have proved that the potassium-argon dates were so wrong?
This is a paper that the creationists love to cite. And yet the citation is out-of-context. That cited paper on page 4603 tells us:
The matrix rock of the Hualalai nodules was erupted during 18011802 [Richter and Murata, 1961] and, indeed, can be said to contain no measurable radiogenic argon within experimental error (Table 2)
What this means is that the lava was, in this particular study, dated to zero age within the margin of error. (This is not surprising since the rate of radioactive decay in potassium-argon method are very slow and the instrumentation was not as good as it is today. Think of using the hour hand of a watch to time a ten-yard dash.)
Thus
the creationist have lied about, or did not understand the contents of this paper.
The two-hundred year old lava was not what scientists were trying to date. What was being studied were xenoliths (also called inclusions). What these are bits of rock that are embedded within the lava flow. These rocks are
older than the lava flow. They were carried up by the magma, but the magma was not hot enough to melt them. Thus one should not be surprised that these bits of rock date older than two centuries old since they are well over two centuries old. Furthermore the study was trying to see in this dating technique is appropriate for xenoliths. They found that it was not. One of the problems which they had was excess argon.
Source