The Food and Drug Administration (FDA) counsels Americans: "Don't change your diet" over Acrylamide worries. The Tufts University Health & Nutrition Letter writes that we've probably been consuming Acrylamide for centuries, "with the human life span only getting longer during that time." And the journal Nature first suggested in 2002 that human beings may be immune to Acrylamide's allegedly carcinogenic effects anyway.

None of this seems to matter to CSPI, whose "experts" are always eager to push Americans toward irrationally fearing their food. CSPI's über-food-nanny Michael Jacobson raised needless Acrylamide alarms last year, and claimed a few months ago (without any proof) that the chemical causes "tens of thousands" of cancers among Canadians. CSPI even colluded with an "environmental" front group for California trial lawyers filing Acrylamide lawsuits against American restaurants and food producers.

Yesterday's CSPI press conference, then, shouldn't have surprised anyone. Jacobson corralled the press to declare that Acrylamide causes "8,900 cancers per year" among Americans, and announced that his group had filed a lengthy petition with the FDA, demanding that the agency force food manufacturers to lower the Acrylamide content of their products.

CSPI is calling for a new regulatory scheme that would force manufacturers within each food category (say, potato chips or freeze-dried coffee) to lower their Acrylamide levels to the product's national average. "The most-contaminated brands," Jacobson told reporters, "should certainly be able to get to the level of half of their competitors." But this seemingly innocuous plea only tells half the story.

In CSPI's actual petition, the group asks the FDA to continuously lower the legal threshold each time the food producers at the top end of the Acrylamide spectrum change their ways. "Thus," the petition demands, "with each iteration the interim acceptable level will fall." And, we might add, American food companies will have to spend more and more money to meet an ever-changing standard for a substance that no one can prove is harmful to humans.

Other highlights from CSPI's petition that didn't make it into the press release:

• CSPI told the FDA that although it doesn't really have "direct regulatory authority over the retail sale of food" in fast-food chains, it should use its authority to regulate "interstate shipment of such food to each retail store" as a legal loophole.
• CSPI claimed that "Acrylamide might cause (pancreatic) cancer in humans," while admitting in a footnote that the authors of their cited study "did not find an association between Acrylamide and cancer."
• In a move reminiscent of its famed "Liquid Candy" and Olestra fiascos, CSPI admits in the text of its petition that it "adjusted [USDA] consumption data" in order to arrive at its estimates of just how much Acrylamide the average American consumes. The result? A 27 percent increase that exists only in CSPI's fertile imagination.
• CSPI also concedes that the data and methods it used to arrive at the 8,900 number are outdated, and that "using more recent EPA methods for projecting cancer-risk findings may result in an estimate several-fold less."
• Showing its legendary disdain for French fries and potato chips, CSPI argues in its petition that Acrylamide is "not an 'inherent natural substance' of a food such as potatoes," since raw tubers' Acrylamide levels are so low. Apparently, CSPI blames Americans for cooking their potatoes.

Perhaps the most glaring flaw in CSPI's petition, however, comes at the very end -- where Michael Jacobson signed his name to certify (as the FDA requires) that his petition "includes representative data and information known to the petitioner that are unfavorable to the petition."

Oops. Jacobson apparently forgot to mention the wealth of science that contradicts his rash position -- to say nothing of the evidence that vegetables like beets and spinach are also Acrylamide-heavy. But spinach and beet farmers don't make easy targets the way Fortune 500 companies do, so we can't expect CSPI to take issue with their offerings.