Type II is entirely preventable. An estimated 90% of type II cases are already completely attributed to lifestyle problems, and while there isn’t completely conclusive evidence yet, I’d bet that a big chunk of the remaining 10% can be laid at the feet of inappropriate diet and lack of effective exercise in the future. The evidence is so overwhelming that you really don’t even need to argue about this one.
Type I is possibly preventable, but the case is much less clear.
Your risk of type I is higher if you have relatives with it, but even identical twins do not always develop the disease, strongly suggesting an environmental or autoimmune trigger. Cow’s milk has been suggested as one strong candidate, as infants who were breast-fed have a much lower incidence of diabetes than infants who drank formula or milk.
You may not buy into all of the paleo diet claims, but there is some clinical and research evidence that some foods are especially problematic for susceptible individuals and can result in the development of autoimmune disorders like vitiligo, alopecia, Crohn’s and celiac, arthritis, and multiple sclerosis, among others. MS may be the most spectacular of these types of diseases since it results in neurological problems. There are two video presentations Cordain on the etiology of MS, TED Talk by Dr. Wahls) about how dietary interventions can help with MS.
In addition to cutting out grains, legumes, and dairy per the standard paleo recommendations, the autoimmune protocol of the diet contraindicates nightshades (potatoes, tomatoes, eggplants), peppers (bell peppers, hot peppers, etc. and spices containing peppers like curry powder, paprika, and chili powder) and eggs. Since there is already an established association in the mainstream medical community with dairy and type I, I don’t think it’s too much of a stretch to suggest that some or all of these other foods might be contributing factors as well. There’s no conclusive evidence at this time, however.
For type I diabetics, a study on a low carbohydrate diet showed that it significantly decreases insulin dependence and prevents large swings in blood glucose levels. A recent study — one of the few so far that specifically looked at a paleo-style diet — shows that a paleo diet is apparently also good for insulin control. This is not a big surprise considering that the food restrictions and recommendations often add up to paleo being a de-facto low glycemic index and/or low carb diet, depending on your food choices day to day.
So, changing your diet might help prevent the development of even type I diabetes, and if you eventually develop it anyway, will help with control of blood sugar swings, reduce your use of insulin, and probably decrease your chances of complications from the disease.
As for the exercise angle, it’s well known that glucose sensitivity changes in response to exercise. This change apparently takes place at the cellular level involving glucose transport molecules. This is quite apart from any body composition changes as an adaptation to exercise, like decreased adiposity and increased lean body mass. Anything you can do to reduce your body fat to appropriate levels will change both insulin signaling and your hormone profile in various ways, and exercise is positively associated with long-term maintenance of a healthy weight.