The development of Proto-Germanic ai in Nordic is very complex. It depends on a number of factors (stressed/unstressed, followed by w, R, h etc.) - see e.g. [Sandøy 2017][1]

PGmc. ai (before w) > ON ǣ  (Heusler 1967: 27)

PGmc. ai > OWN ei, OEN ē ([Versloot 2017:][2] 285)

PGmc. ai (before non-rhotacized R and also h) > ON ā (this ā further developed into ē in i-umlaut contexts)

PGmc. ai was lost medially in NGmc ([Stiles 2017:][3] 911)


  [1]: https://www.degruyter.com/view/book/9783110197068/10.1515/9783110197068-084.xml
  [2]: https://www.degruyter.com/view/journals/flin/51/s38/article-p281.xml
  [3]: https://www.degruyter.com/view/book/9783110523874/10.1515/9783110523874-009.xml