Recent government reports on the failure of American airport screeners to detect threat objects at security checkpoints may provide ammunition for proponents of the controversial Computer Assisted Passenger Prescreening System (CAPPS II) database solution, which is currently stalled by myriad snafus too numerous to mention.

Human error

The Department of Homeland Security (DHS) Inspector General and the Congressional General Accounting Office (GAO) have both submitted reports on the competence of airport passenger and baggage screeners, and found, not surprisingly, that they are no more effective today than they were before the security frenzy brought about by the 11 September atrocities.

In testimony before the House Aviation Subcommittee, Inspector General for Homeland Security Clark Ervin and GAO Managing Director Norman Rabkin said that the Transportation Security Administration's (TSA's) well-paid screening personnel are no more effective than the inexpensive rent-a-cops provided by private contractors. A comparison between federal screeners and those participating in a pilot program for private contractors called the PP5 Program.

According to Ervin, federal and private screeners "performed about the same, which is to say, equally poorly." He added that "this result was not unexpected, considering the degree of TSA involvement in hiring, deploying, and training the [private sector] screeners."

It's believed that TSA's interference in the PP5 Program and its bureaucratic inertia are important reasons why the private-sector screeners failed to outdo their civil-service counterparts. Both reports are biased against the TSA. They assume that TSA is a lost cause, although, ironically, it had originally been touted as a much-needed fix for the incompetence of private contractors, upon whom blame for the 9/11 atrocity was conveniently fixed in the immediate aftermath.

It now appears that TSA is seen as the chief source of security incompetence and failure. "TSA's tight controls over the pilot program restricted flexibility and innovation that the contractors may have implemented to perform at levels exceeding that of the federal workforce. TSA needs to establish a more robust pilot program that allows greater flexibility to test new innovations and approaches," Ervin said.

Defective detectives

Indeed, passenger screening is no better than it was 17 years ago. Covert testing conducted in 1978 - back when screeners were reasonably polite and quick and unobtrusive about their business - found that 13 per cent of threat objects passed undetected. Today, in the wake of post-9/11 security hysteria, and its attendant aggressive bullying of the public and punishment-strip searching of anyone daring to pass a sarcastic comment, the figure is 20 per cent.

TSA Administrator Admiral David Stone defended his outfit and took issue with the reports. "Testing in the Nineties was in no way even comparable to what we do," he said.

While it may be true that today's covert testing is more sophisticated, detection equipment has also improved to make the screeners' jobs easier, though he neglected to emphasize this fact. The red teams and the blue teams have both got better tricks up their sleeves, so there's certainly nothing unfair about the penetration tests, as Stone tried to imply. Still, bad news for human screeners may well be good news for technology.

Database Hell

Stone showed little enthusiasm for the PP5 Program, but he is a big proponent of CAPPS II, having touted it before the same House committee back in March as a scheme promising to deliver "vital impact ... on aviation security."

He has studied the vendor's PR boilerplate with great care. CAPPS II is a "second-generation prescreening system [that] will be a centralized, automated, threat-based, real time, risk assessment platform ... expected to employ technology and data analysis techniques to conduct an information-based identity authentication," he gushed.

The system is a product of aviation defense contractor Lockheed Martin Corporation, promoted by US Transportation Secretary and former Lockheed Martin Vice President Norman Mineta, Stone's boss.

At present, the grand "risk assessment platform" is mired in failure. What little of it currently works has not been tested adequately because carriers are withholding passenger data in fear of a public backlash on privacy grounds.

The accuracy of the many databases that CAPPS II will scour for its incriminating evidence has not yet been established. Procedures for passengers to detect inaccurate data, and get inaccurate data and false positives resolved, have not been implemented. Major privacy threats inherent in the system, particularly those involving restrictions on access, have not been addressed. The potential for malevolent identity thieves to impersonate innocent travelers remains high.

False consciousness

This is all good, because CAPPS II is one of the worst possible solutions to airport security. It won't prevent terrorists from flying; rather, it will increase the probability of another successful attack using commercial aircraft.

The reason is painfully obvious: a group can very conveniently use the system to pre-screen its members and discover which of them have profiles that result in extra scrutiny. Thus CAPPS II is a superb tool for terrorists to use in assessing airport defenses. A group of unarmed terrorists can board two or three flights in succession and observe how the system reacts to them. If, after a few trial runs, they discover that they're allowed to board unchallenged, they can assume that their profiles do not trigger a warning. Armed with that information, they'll stand a good chance of mounting a real attack.

CAPPS II is a disaster for two reasons: first, it will create a false sense of security among airline staff and provide further excuses for screeners to perform poorly; and second, it offers terrorists an excellent training device that they can use to assemble a group of people who can get onto airplanes without arousing suspicion. Ironically, the closer CAPPS II comes to achieving its stated goals, the more effective it will become as a terrorist tool.

So it is indeed good that its development is going poorly. The problem, however, is that the recently publicized failures among human screeners will provide rationale to rush it into service. CAPPS II may well find itself on a fast track, pushed hard by those who would exploit the popular misconception that computers and other high-tech gizmos can compensate for human fallibility. ®

Thomas C Greene is the author of Computer Security for the Home and Small Office, a complete guide to system hardening, online anonymity, encryption, and data hygiene for Windows and Linux, available at discount in the USA, and in the UK.

Related stories

American Airlines data used to test passenger snoop system The wrong stuff: what it takes to be a TSA terror suspect

Campaigners fight biometric passports
Data on 10m Northwest fliers handed to NASA for testing
US using EU airline data to test CAPPS II snoop system
Commission agrees US access to EU citizen personal data
Congress threatens two hi-tech Gestapo programs