NARRATION: In one of his first moves as attorney general, Jeff Sessions won’t renew a scientific commission that was working to make analysis of crime scene evidence more reliable.

Why is that move so controversial? Take the story of microscopic hair analysis, a forensic technique that for decades was said to be nearly foolproof, and was used to send thousands of people to prison.

MAX HOUCK (PHYSICAL SCIENTIST, F.B.I. LABORATORY, 1994-2001): We would look at hairs all the time, every day. If you look at something every day, routinely, you get very good at noticing small differences.

NARRATION: Max Houck was one of the F.B.I.’s elite hair and fiber experts, who used microscopes to compare hair from a crime scene to hair from a suspect.

MAX HOUCK: Some of the people that I worked with were fantastic. They had such a keen eye, such good discrimination. They saw things that I just didn’t think you could see.

NARRATION: But by the mid 90s, DNA analysis was available and would challenge those experts. Houck remembers the day when two hairs that the F.B.I. believed came from the same person were sent for an additional DNA test. The results were jarring.

MAX HOUCK: The hairs didn’t come from the same individual. We started talking about, well, what did we see? Why did we think it did match? How good are we, in a sense.

NARRATION: Houck used D.N.A. to check how often hair samples that looked the same under a microscope actually came from different people, and in 2002, published his results.

MAX HOUCK: About 11 percent of the time, mitochondrial D.N.A. said, no, that hair actually came from someone else. It kinda shook us up.

NARRATION: Today, the results are stark: Of the more than 300 people who’ve been exonerated by DNA, one-fifth of their convictions involved faulty hair analysis.

The technique came under scrutiny after three high-profile cases  in 2012.

PETER NEUFELD (CO-FOUNDER, INNOCENCE PROJECT): Three different crimes, three different men, all exonerated. Three different F.B.I. examiners. In fact, in one of the three cases, it was actually the chair of the F.B.I. hair unit. That’s when the F.B.I. began to really worry.

ARCHIVAL:
PETE WILLIAMS: The F.B.I. says it is now going through thousands of other cases from the days before D.N.A. testing to see whether witnesses or prosecutors exaggerated the significance of the F.B.I.’s hair analysis.

NARRATION: Initial results from that landmark review of thousands of cases found that the F.B.I. overstated the importance of its evidence with alarming frequency.

PETER NEUFELD: I fully expect that in 90 to 95 percent of those cases, the F.B.I. will conclude that their own agents provided scientifically invalid, erroneous testimony.

NARRATION: While not all those cases necessarily would have been overturned, Attorney General Sessions is suspending the hair analysis review, along with the forensic science commission. And critics say that undermines efforts to make crime scene analysis more accurate and keep innocent people out of prison.

(END)