After the small plates of Nephi, things are different. In the Book of Mosiah, there are no occurrences of "I, Benjamin," "I, Mosiah," or "I, Limhi," but Zeniff uses that form twice in his account. The large Book of Alma has "I, Alma" only four times. That book, not the Book of Helaman, is where we find the only two occurrences of "I, Helaman" in an epistle. Throughout the Book of Mormon, both Mormon and Moroni identify themselves in that form 16 and 17 times, respectively, most of which are done in the role of editors to make it clear who is speaking. The phrase "I, Ether" does not occur at all.
It seems that the person who raised the objection based on "I, Nephi" noted this difference in style between Nephi's writings and the other parts of the Book of Mormon. I think because it struck him as an "obvious" difference, it seemed "made up" - and thus evidence of fraud. There may be something of a circular argument here, or a catch-22 from Book of Mormon critics. If the style of two writers seems the same, it shows that there was one author, making the book a fraud. If the style seems obviously different, it's evidence that style differences were just made up, making the book a fraud.
Can only unnoticeable differences count in favor of multiple authorship? If so, once they are pointed out, I suppose they would no longer qualify since they have been noticed and are now "obvious." Actually, the importance of subtle, hard-to-notice and hard-to-fake differences is the premise in the computerized wordprint analysis done a number of years ago by John Hilton (using improved methods relative to an earlier effort known as the Larsen study). See John L. Hilton, "On Verifying Wordprint Studies: Book of Mormon Authorship," BYU Studies, 30 (Summer 1990):89-108. An analysis of subtle patterns in language - things that are very difficult to consciously control - can be used to identify characteristic patterns of different authors and to statistically distinguish authors. Based on that work, there is good reason to believe that there were multiple authors of the Book of Mormon, and that these authors were almost certainly not Joseph Smith, Oliver Cowdery, or Solomon Spaulding. (BTW, my wife's masters' thesis in statistics was a wordprint study advised by Dr. Hilton on the writings of Paul.)
This issue of "I, Nephi" shows a minor but important difference between the books from the small plates of Nephi and the rest of the text. For example, Brant Gardner's Multidimensional Commentary on the Book of Mormon for First Nephi 1 has this to say about the opening verse that begins with "I, Nephi":
Hugh Nibley first identified 1 Nephi 1:1-3 as a colophon, a structured and typical identificatory passage used at the beginning or end of many ancient documents (Nibley Since Cumorah, 1967, pp. 170-171). The essential elements are the identification of the writer, the writer's lineage, and at times a statement of the veracity or trustworthiness of the written text. . . .I hope that provides some helpful insight into the "I, Nephi" issue and the complex styles and structures of the text.
The hallmark of the colophon is the personal introduction of the material by the writer. While Nephi's introduction is clearly the most formal, the introduction to the written text by the writer continues for most of the material from Nephi to the end of Omni (Jacob's personal introduction is perhaps the least formal, and the furthest from the structures of a colophon). Once the Book of Mormon picks up with Mosiah, however, the personal introductions cease, and are replaced by a typically chronological introduction (Alma 1:1 "Now it came to pass that in the first year of the reign of the judges...."; Helaman 1:1 "And now behold, it comae to pass in the commencement of the fortieth year of the reign of the judges..."; 4 Nephi 1:1 "And it came to pass that the thirty and fourth year passed away,..."). None of these qualifies as examples of colophons, and even the personal introductions lack the formulaic precision of Nephi's introduction.
The first clear division can be made between the personalized introductions of the 1 Nephi through Omni material, and all books which follow. This division is precisely that between the small plate material and the large plate material. The small plates were written in the first person, and the large plates were abridged. More will be said later about the structure and content of these different sources, but for now it is sufficient that the introductory material for the books in each section is clearly different, and follows a different literary imperative.
So, if I may ask, are such stylistic differences evidence that the Book of Mormon was just made up to appear to have different authors? And are the more subtle, less-obvious differences detected by computer analysis of patterns involving minor words also evidence that the book was made up? Is the presence of sophisticated Hebraic patterns like chiasmus also evidence of fraud? Really, I've encountered a few examples where critics point to Semitic elements in the text as examples of fraudulent borrowing from the Bible, and when they don't find enough Semitic elements (like details on how the Law of Moses was observed), they find that as evidence of fraud as well. To me, the zealous effort to reject the Book of Mormon no matter what represents a modern "marvelous work and a wonder," though I question the inspiration behind it.