According to Ancestor Search, OneWorldTree was discontinued in late 2013. This seems to be the case, as there appears to be no way to limit searches at Ancestry.com to OneWorldTree, and it does not show up as a distinct category in the search filter.
This was a user-contributed tree on Ancestry that attempted to merge everyone's data into "one" tree. However, the matching mechanisms were (for understandable reasons) designed loosely, and errors were rampant. Correcting errors was very difficult, because the system essentially allowed users only to "vote" for the correct information. Removing incorrect information was difficult, and there was no way to add notes or sources to explain why one set of information was more reliable than another.
The reliability of OneWorldTree was worse than the Ancestral File, for a couple of reasons. Firstly, its primary source of data was the internet, and since it is so easy for people to upload information to the internet, a significant volume of "bad" information was replicated into OneWorldTree.
Secondly, at least in its early days, the algorithms used for matching individuals did not acknowledge some basic genealogical facts, such as that it was common for parents whose child had died to name another child with the same name. As a result, several families had one record representing 2, 3 or even more distinct individuals. The algorithm (and subsequent user voting) would then determine which of the various birth and death dates to display, resulting in individuals being displayed with birth dates after death dates and similar impossibilities. Similarly, the application of the merging rules in a step-wise fashion across a number of related (and unrelated) people resulted in people being linked to supposed children who were born 40 years before they were. While the Ancestral File also contains similar errors, OneWorldTree took perfectly good data and managed to mangle it with automation, speeding up the process of turning good data into bad. These problems were reported to Ancestry in 2005 and possibly were partially rectified since then (it is now impossible to check whether early incorrect merges were ever corrected), but any information whose only source is OneWorldTree should be treated with a high degree of suspicion (possibly worse than no source at all).