Opportunists call for weakening encryption in wake of Paris attack

Creating coding 'backdoors' to help law enforcement stop terrorists is fraught with complexities

Share on FacebookShare on Google+Tweet about this on TwitterShare on LinkedInEmail this to someonePrint this page

It’s nat­ur­al to look for a scape­goat after some­thing ter­ri­ble happens.

Some are now argu­ing that if only we could have read encrypt­ed com­mu­ni­ca­tions, per­haps the Paris ter­ror­ist attacks could have been stopped. … Wrong.

Read every sto­ry you see about Paris care­ful­ly and look for evi­dence that encryp­tion played a role. You won’t find it.

More: Keep your clients’ infor­ma­tion secure and employ­ees safe

There’s a rea­son The Patri­ot Act was passed only a few weeks after 9/11—and it wasn’t because Con­gress final­ly was able to act quick­ly and effi­cient­ly on some­thing. The speed came because many ele­ments of the Patri­ot Act already had been writ­ten, and forces with an agen­da were sit­ting in wait for a dis­as­ter so they could push that agen­da. That is wrong.

So here we are, once again, faced with polit­i­cal oppor­tunism after an unthink­able human tragedy, and we must remain strong in the face of it. There is no sim­ple answer to ter­ror­ism, and we should all know this by now.

Beware knee-jerk reactions

And so there must be no sim­ple dis­cus­sion about the use of encryp­tion in the West­ern world. The debate requires care­ful analy­sis, and we owe it to every­one who ever died for a free soci­ety to con­sid­er this thoughtfully.

The basics are this: Only recent­ly, com­put­ing pow­er has become inex­pen­sive enough that ordi­nary cit­i­zens can scram­ble mes­sages so effec­tive­ly that even gov­ern­ments with near-infi­nite resources can­not crack them. Such secret-keep­ing pow­ers scare gov­ern­ment offi­cials, and for good rea­son. They can, the­o­ret­i­cal­ly, allow crim­i­nals and ter­ror­ists to com­mu­ni­cate with a cloak of invisibility.

Not sur­pris­ing­ly, sev­er­al gov­ern­ment offi­cials have called for a method that would allow law enforce­ment to crack these codes. There are many schemes for this, but they all boil down to some­thing akin to cre­at­ing a mas­ter key that would be gen­er­at­ed by encryp­tion-mak­ing firms and giv­en to gov­ern­ment offi­cials, who would use the key only after a judge grant­ed per­mis­sion. This is some­times referred to as cre­at­ing “back­doors” for law enforcement.

sh_endpoint encryption_750Gov­ern­ments already can lis­ten in on tele­phone con­ver­sa­tions after obtain­ing the prop­er court order. What’s the dif­fer­ence with a mas­ter encryp­tion key?

Sad­ly, it’s not so simple.

For starters, U.S. firms that sell prod­ucts using encryp­tion would cre­ate back­doors, if forced by law. But prod­ucts cre­at­ed out­side the Unit­ed States? They’d cre­ate back­doors only if their gov­ern­ments required it. You see where I’m going. There will be no glob­al mas­ter key law that all cor­po­ra­tions adhere to.

By now I’m sure you’ve real­ized that such laws would only work to the extent that they are obeyed. Plen­ty of com­pa­nies would cre­ate rogue encryp­tion prod­ucts, now that the mar­ket for them would explode. And, of course, ter­ror­ists are hard at work cre­at­ing their own encryp­tion schemes.

There’s also the prob­lem of exist­ing prod­ucts, cre­at­ed before such a law. These have no back­doors and could still be used. You might think of this as the genie out of the bot­tle prob­lem, which is real. It’s very, very hard to undo a tech­no­log­i­cal advance.

Dan­gers of dis­man­tling encryption

Mean­while, cre­ation of back­doors would make us all less safe. Would you trust gov­ern­ments to store and pro­tect such a mas­ter key? Man­ag­ing defense of such a uni­ver­sal secret-killer is the stuff of movie plots. No, the mas­ter key would most like­ly get out, or the back­door would be hacked. That would mean ille­gal actors would still have encryp­tion that worked, but the rest of us would not. We would be fight­ing with one hand behind our backs.

In the end, it’s a famil­iar argu­ment: dis­abling encryp­tion would only stop peo­ple from using it legal­ly. Crim­i­nals and ter­ror­ists would still use it illegally.

Is there some cre­ative tech­no­log­i­cal solu­tion that might help law enforce­ment find ter­ror­ists with­out destroy­ing the entire con­cept of encryp­tion? Per­haps, and I’d be all ears. I haven’t heard it yet.

Only a few weeks after 9/11, a soft­ware engi­neer who told me he was work­ing for the FBI con­tact­ed me and told me he was help­ing cre­ate a piece of soft­ware called Mag­ic Lantern. It was a type of com­put­er virus, a Tro­jan horse key log­ger, that could be remote­ly installed on a target’s com­put­er and steal pass phras­es used to open up encrypt­ed documents.

The pro­gram­mer was uncom­fort­able with the work and want­ed to expose it. I wrote the sto­ry for msnbc.com, and after deny­ing the exis­tence of Mag­ic Lantern for a while, the FBI ulti­mate­ly con­ced­ed using this strat­e­gy. While we could debate the mer­its of Mag­ic Lantern, at least it con­sti­tut­ed a tar­get­ed investigation—something far, far removed from ren­der­ing all encryp­tion ineffective.

For a far more detailed exam­i­na­tion of these issues, you should read Kim Zetter at Wired, as I always do. Then make up your own mind.

Don’t let a politi­cian or a law enforce­ment offi­cial with an agen­da make it for you. Most of all, don’t allow some­one who cap­i­tal­izes on tragedy mere hours after the first blood is spilled—an act so crass it dis­qual­i­fies any argu­ment such a per­son makes—to influ­ence your thinking.

More on encryption:
Encryp­tion must be strong, used prop­er­ly to reli­ably pro­tect data
Anthem breach shows need for wider encryp­tion of sen­si­tive data
Let’s Encrypt’ seeks to fos­ter trust in web traffic


Posted in Cybersecurity, Data Security, Featured Story