The Groundwork and the good will
While the Groundwork gives pre-eminent place to the good will (guten Wille), it says nothing so far as I know about the bad or evil will. Of course, it doesn't logically follow that if there is a good will there is also a bad or evil will: apart from the good will there might only be a non-good or neutral will or indeed no other kind of will. More than that, Kant talks of the good will at the conceptual level, not that of reality. He says only that we cannot think of anything in the world which is good without limitation - qualification - except the good will. Given the criticism of the ontological argument in the Critique of Pure Reason (KrV A592 = B620 ff.), Kant would be the last to argue from concept to reality. So the good will may not be instantiated and there may be no good will.
The reality of the bad will
Yet the evidence is clear, if not in the Groundwork, that Kant does believe that there is an evil will. The data for this belong to Religion within the boundaries of mere reason (1792-3; 2nd ed. 1794). Kant supposes that we not only have the concept of a bad or evil will but that such a will is inherent in humankind – and this is a reality claim.
The real clue to the bad will is not in the Groundwork
It's taking you beyond your reading but Kant's Religion within the Boundaries of Mere Reason is far more illuminating about the bad will - it could hardly be less - than the Groundwork. But the account does link back to the Groundwork as the references to the categorical imperative and maxims, highlighted below, make clear.
Of all his texts in the area of practical philosophy, Kant's Religion within the Boundaries of Mere Reason has long been widely held to be the least satisfactory. Quite at odds in tone with the bulk of his earlier work on the subject, in which he champions human freedom and exalts the dignity of rational agency and the nobility of the good will's commitment to morality, the Religion paints a dark portrait of humanity as a race perpetually in bondage to its own sinful nature; its central claim being that all human beings possess a radical propensity to evil that makes evil deeds inevitable for us. The book has been attracting controversy ever since its publication, which notoriously was greeted with dismay by many of those who had previously been most enthusiastic in their praise of him. Goethe and Schiller, for example, both viewed the claims of the Religion as arbitrary and unjustified concessions to pre-Enlightenment prejudice in a particularly pernicious form, that of a broadly Augustinian theology of original sin, concessions entirely unmotivated by any consideration internal to the basic Kantian project. Nor, by and large, have more recent commentators been any more sympathetic. Their criticisms tend to be built less on a visceral opposition to the encroachments of an allegedly misanthropic Christianity, resting instead on a careful analysis of the particular claims Kant makes throughout the text. (Seiriol Morgan, ‘The Missing Formal Proof of Humanity's Radical Evil in Kant's "Religion"’, The Philosophical Review, Vol. 114, No. 1 (Jan., 2005), pp. 63-114: 63.)
we are all … subject to a propensity (Hang) to evil. Unfortunately, it cannot be said that Kant is particularly clear on what exactly such a propensity might be, defining propensity cryptically as "the subjective ground of the possibility of an inclination, insofar as this possibility is contingent for humanity in general" (R 6:29) before notoriously going on to illustrate the idea with the example of the alleged propensity of all "savage" people for intoxicants. This is quite unhelpful, since such a propensity would be the disposition to develop an ongoing desire for a substance upon experiencing its intoxicating effect for the first time. Evil however cannot lie in the inclinations, for then it would be a function of the causal laws of the world and not free choice, and consequently would not be imputable (R 6:21) … What does seem clear is that Kant is claiming the following. The propensity to evil differs from a predisposition because even though it is innate, it is appropriately thought of as having been brought by the human being on herself (R 6:29). It is the ground of the possibility of our adopting specific evil maxims (ibid.). It is universal and "woven into human nature" (R 6:30), but nevertheless is rooted in the will's free choice, so despite being subjectively necessary for everyone, it must be thought of as an accidental property of the human being (R 6:32). The propensity is evil in itself (R 6:37), and it is inextirpable (R 6:31), but since it is rooted in our freedom, it must be possible for freedom to overcome it (R 6:37). In human action, the propensity manifests itself in three different degrees:** frailty, impurity, and depravity** (R 6:29-30). Kant takes the apostle Paul to be describing the phenomenon of frailty* [*akrasia or weakness of will: GT] when he reported that "I do not do what I want, but I do the very thing I hate" (Rom 7:15). An individual manifesting frailty is someone who has adopted a good **maxim, but finds himself acting counter to it because the incentive provided by the moral law that his will endorses is somehow weaker than that provided by cajoling inclination. In the case of impurity, once again the agent's fundamental orientation is good, in that his will endorses a maxim that aims at compliance with the moral law. But for such a person, the moral incentive is insufficient to determine the will by itself, and so in order to actually perform an action required by the Categorical Imperative, the agent needs some additional motivation from the inclinations, and so his performance of the right action relies on heteronomous considerations. Such an action would be an impure action; an agent's heart is impure when he frequently or invariably requires this self-centered stiffening of his supposedly moral resolve. Finally, in depravity or corruption, the individual's descent into evil is complete. Here the maxim is not good at all, since the corrupt person has made a perverse choice to reverse the appropriate priority of the incentives, by subordinating the moral law to her own self-love, and simply acts accordingly. Kant characterizes the first two grades of the propensity to evil as unintentional guilt, but the final stage is clearly deliberate (if self-deceptive) guilt (R 6:38), and consequently those who perversely subordinate the moral maxim to that of self-love are to be designated evil people (R 6:30). (Seiriol Morgan: 66-8.)
The will of such people is bad without limitation or qualification in Kant’s view - bad or evil in itself. A significant problem is, however, how Kant can make good his claim about the real existence of evil – of the bad or evil will.
Perhaps the most obviously problematic aspect of the text is the yawning gap at the heart of the argument, at the point where solid argument is most needed. For, lacking as he obviously does empirical acquaintance with the behavior of each and every human being past, present, and future, Kant will need to present an a priori argument in order to earn the right to assert that all human beings have such a propensity. Indeed, since he thinks that this characteristic is necessary (R 6:32), even if he were per impossibile to have such an acquaintance, this would still not underwrite the modality of the claim. There are certainly various indications in the text that Kant is perfectly aware that some kind of formal argument will be needed to establish his position. For instance, at the beginning of part 1, he states that in order to call a human being evil, we must be able to infer a priori from her actions an evil maxim, and consequently an evil common grounding of all her particular evil maxims (R 6:20). And on pages 32 to 33 of the Prussian Academy Edition of the Religion, he clearly implies that such a proof is available. But famously he does so in the course of relieving himself of the burden of providing it, "in view of the multitude of woeful examples that the experience of human deeds parades before us." Two pages later, he tells us that the existence of the propensity to evil in human nature "can be established through experiential demonstrations of the actual resistance in time of the human power of choice against the law" (ibid., 35). But as many have indignantly pointed out, it cannot. All that the woeful parade of human deeds can show us is that there are evil people, or at most, that evil is common and widespread. It would be an entirely reckless generalization to conclude from the undeniably extensive litany of the crimes that human beings have carried out that every single human being has a propensity to evil, and indeed actually is evil, and that the root of this evil lies in human nature. Furthermore, it is clearly a central tenet of the Critique of Pure Reason that transcendental propositions such as this cannot be confirmed by appeal to examples from experience (KrV A 554/B 582). So his claims are crying out for a transcendental deduction that he does not provide. Obviously the suspicion has to be that he does not do so because he has no such argument to hand. (Seiriol Morgan: 64-5.)
Seiriol Morgan, ‘The Missing Formal Proof of Humanity's Radical Evil in Kant's "Religion"’, The Philosophical Review, Vol. 114, No. 1 (Jan., 2005), pp. 63-114.
Kant, Immanuel ; Gregor, Mary [Ed], Groundwork of the Metaphysics of Morals, ISBN 10: 0521626951 / ISBN 13: 9780521626958
Published by Cambridge University Press, Cambridge University Press.
Kant, Immanuel; Wood, A. et all [Edd.], Religion Within the Boundaries of Mere Reason, ISBN 10: 1316604020 / ISBN 13: 9781316604021
Published by Cambridge University Press 2018.
Kant, Immanuel, Critique of Pure Reason, Translated and edited by Paul Guyer and Allen W. Wood. ISBN 10: 0521657296 / ISBN 13: 9780521657297.
Published by Cambridge University Press, 1998.
There is also relevant material in Bk III of:
Kant, Immanuel/ Louden, Robert B. (Translator), Anthropology from a Pragmatic Point of View, ISBN 10: 0521671655 / ISBN 13: 9780521671651
Published by Cambridge University Press, 2006. [Original 1798.]