"[84] ... For reasons already discussed in relation to husbands and wives, a bank cannot be expected to probe the emotional relationship between two individuals, whoever they may be. Nor is it desirable that a bank should attempt this. Take the case where a father puts forward his daughter as a surety for his business overdraft. A bank should not be called upon to evaluate highly personal matters such as the degree of trust and confidence existing between the father with his daughter, with the bank put on inquiry in one case and not in another. As with wives, so with daughters, whether a bank is put on inquiry should not depend on the degree of trust and confidence the particular daughter places in her father in relation to financial matters. Moreover, as with wives, so with other relationships, the test of what puts a bank on inquiry should be simple, clear and easy to apply in widely varying circumstances. This suggests that, in the case of a father and daughter, knowledge by the bank of the relationship of father and daughter should suffice to put the bank on inquiry. When the bank knows of the relationship, it must then take reasonable steps to ensure the daughter knows what she is letting herself into.
85. The relationship of parent and child is one of the relationships where the law irrebuttably presumes the existence of trust and confidence. Rightly, this has already been rejected as the boundary of the O'Brien principle. O'Brien was a husband-wife case. The responsibilities of creditors were enunciated in a case where the law makes no presumption of the existence of trust and confidence.
86. But the law cannot stop at this point, with banks on inquiry only in cases where the debtor and guarantor have a sexual relationship or the relationship is one where the law presumes the existence of trust and confidence. That would be an arbitrary boundary, and the law has already moved beyond this, in the decision in Burch. As noted earlier, the reality of life is that relationships in which undue influence can be exercised are infinitely various. They cannot be exhaustively defined. Nor is it possible to produce a comprehensive list of relationships where there is a substantial risk of the exercise of undue influence, all others being excluded from the ambit of the O'Brien (Barclays Bank Plc v O'Brien [1993] UKHL 6; [1994] 1 AC 180) principle. Human affairs do not lend themselves to categorisations of this sort. The older generation of a family may exercise undue influence over a younger member, as in parent-child cases such a as Bainbrigge v Browne, 18 Ch D 188 and Powell v Powell [1900] 1 Ch 243. Sometimes it is the other way round, as with a nephew and his elderly aunt in Inche Noriah v Shaik Allie Bin Omar [1929] AC 127. An employer may take advantage of his employee, as in Credit Lyonnais Bank Nederland NV v Burch [1996] EWCA Civ 1292; [1997] 1 All ER 144. But it may be the other way round, with an employee taking advantage of her employer, as happened with the secretary-companion and her elderly employer in In re Craig, Decd [1971] Ch 95. The list could go on.
87. These considerations point forcibly to the conclusion that there is no rational cut-off point, with certain types of relationship being susceptible to the O'Brien principle and others not."