In June 2006, the International Corporation for Assigned Names and Numbers (ICANN) approved a plan that would allow for the creation of hundreds of new domain names. The Internet as we know it currently operates using 12 generic top-level domain names (gTLDs) to direct traffic. Some of the more common gTLDs include .com, .net, .edu, .org, and .gov. ICANN's proposed changes would make it possible for applicants to create entirely new gTLDs, anywhere from 3 to 63 characters long and in nearly any alphabet, such as Arabic or Chinese. ICANN released the draft plan for an open comment period, which ran from October 24 – December 15, 2008. The response from the public has generally been negative and it remains to be seen how, if at all, that response will impact the final version of the plan. In whatever form the plan does proceed, however, it presents both exciting opportunities and treacherous pitfalls for trademark owners.

A persistent problem for trademark owners in cyberspace is the conflict between similar but equally valid marks. Until now, the coveted .com domain has gone in a first-come, first-served fashion; where the first to arrive has a valid right to use the mark, those arriving late to the party have been forced to settle for some variation of their own mark. A frequently cited example is that of delta.com. The trademark "Delta" is validly held by an airline, a faucet company, and a dental insurance company, but only one of them can own delta.com (in this example, the airline was the winner). Under the new system, Delta Dental could purchase .delta, or perhaps join a consortium and go in together to create .health or .insurance. Options are virtually limitless under the proposed plan. As the e-marketplace becomes increasingly crowded, the demand for more space becomes correspondingly more urgent. This .com crowding was a major impetus for the creation of the plan in the first place. Moreover, owning and maintaining a unique domain would help businesses whose customers are regularly targeted by scammers and phishers. Suppose Bank of America were to purchase .bofa or .bankofamerica. Customers would be able to feel secure in the certainty that communications received from those domains were authentic. That resulting sense of security could go a long way in adding value and cementing brand loyalty.

However, the plan also presents a minefield for the unwary mark holder. Trademark owners are already required to be hyper-vigilant in defending their marks against infringement, cyber-squatting, and other sources of online consumer confusion. Under this regime, the burden of policing not only remains on the holder of the mark, but will likely increase exponentially. With every new gTLD comes the potential for drastically higher incidences of infringing and/or confusing domain names. While ICANN's rules prohibit infringement, there is no system in place to screen proposed domains for infringement or to verify ownership of a particular mark. Mark owners will be responsible for monitoring new applications and must assert timely objections when they identify potentially infringing new domain names. Incidentally, each objection-related filing comes with a fee of $5,000, both to make an objection as well as to defend against one. Furthermore, organizations that can afford to do so may feel compelled to purchase new gTLDs that they don't particularly need or want, as a preemptive move against competitors and would-be cyber-squatters. This would require substantial expenditures of both time and money, and with the minimal restrictions on new domain names, it would be impossible to guard against all potential infringements.

The implications for trademark owners, costs (the initial application fee alone is $185,000), and an exhaustive review process have raised concerns among some commentators. Negative comments made during the plan's open comment period also indicate that the final plan going forward may end up looking different than the one originally approved. The time designated to begin accepting applications (mid-2009) will likely be delayed. The current internet model, however, is arguably becoming outdated and, at some point, changes will have to be made.