The work only finishes when sensitive information leaves
In September, a senior Morgan Stanley banker covering financial sponsors in Asia attached the wrong file to their weekly client update. That file turned out to be the firm's internal deal pipeline. According to Reuters, by the time the email was retracted, it had already reached recipients outside the firm, and regulators in more than one jurisdiction have since reportedly begun asking questions. The recipients were clients who receive an update every week, which made the incident an ordinary send gone wrong rather than a malicious act. The fact that this was a mistake is what makes it instructive: from legal work to pharmaceutical research, sharing sensitive data with people outside the organization is how business gets done, and each routine send carries some version of the same exposure.
In fact, most of this sharing still happens by email. In a 2025 survey of 461 cybersecurity professionals, Kiteworks reported that 90% of sensitive business communications still travel that way. Because that sharing happens constantly and under deadline, it also happens when conditions are least favorable for careful judgment, and a wrong attachment or recipient is enough to turn a routine send into a disclosure that the organization must explain to its clients and regulators. These are situations where security leaders cannot prevent sensitive information from leaving the organization's four walls, since the business depends on it. This means the question that matters is how much control the organization keeps once the information has gone, which is the core of secure email sharing.
The exposure sits outside your walls and inside your workflow
The Kiteworks survey also attributed 60% of incidents to employee mistakes, which suggests that exposure often begins with a routine send by a legitimate user. The Morgan Stanley case fits that pattern. Every sensitive attachment sent to a partner becomes subject to that partner's security practices, and every send depends on a person choosing the right recipient and the right file at speed. The organization controls neither condition once the message has left, and each additional send extends its exposure into environments it cannot see.
Organizations are not skipping data controls, however, and they have invested in tools such as data security posture management (DSPM) and data loss prevention (DLP). Those investments answer an important question, which is whether something sensitive is about to leave, but they say far less about what happens afterward, when the information is out of the organization's hands and still carries its full value.
Warn, block, or let it go: three responses that each pass the risk along
When a DLP policy detects sensitive content in an outgoing email, it typically offers three responses, and the stakes rise with each one because each response shifts the risk elsewhere.
1. Warn the sender, and the risk moves to the employee
A warning asks the employee to confirm the send. When the recipient is legitimate, the employee has every reason to confirm and continue, so the data leaves unprotected and the outcome rests on the employee's judgment. Over time, a warning that is nearly always confirmed carries less weight, and the organization ends up depending on individual judgment.
2. Block the email, and the risk moves to the deadline
Blocking stops the exposure, but it does not remove the need to share. A contract still has to reach opposing counsel and a dataset still has to reach its research partner. In situations like this, the employee raises an exception, looks for another route, or waits, all of which can delay work that the business has already committed to.
3. Let it go, and the risk moves to someone else's mailbox
Allowing the send keeps work moving, and for routine sharing it is often the right decision. Once the email is delivered, however, the content sits in the recipient's inbox, under the recipient's security. The sensitive information can travel with every forward, reply, and download that follows, which means control shifts to people and systems that the sender does not manage and may never see.
Each of these responses is reasonable on its own terms, but together they show a pattern in which the organization either gives up protection to keep the work moving or gives up progress to keep the data safe. Where that trade-off remains unresolved, employees end up making it themselves, one deadline at a time.
What changes when protection travels with the content
The more useful question is what would change if the secure option were also the easiest one for employees to choose. A secure path that's as easy as Send would mean that following policy no longer competes with meeting a deadline, which removes the trade-off that each of the three responses above forces on the business.
That path depends on where protection lives, because securing the inbox, the portal, or the recipient's environment means relying on systems that the sender does not control. When protection stays with the content itself, however, the destination matters less, since the organization keeps control over who can access the information after it has been sent, no matter which mailbox it reaches.
For security leaders, the effect reaches well beyond a single email. It means fewer exception requests to review, fewer delayed deals and submissions, fewer reasons for employees to build their own workarounds, and an audit trail that extends past the moment of sending, showing who accessed sensitive information and when. Over time, that is the difference between explaining an exposure after the fact and being able to show, on request, how the information was handled.
Introducing Confidencial for Outlook
Confidencial for Outlook brings that protection into the email client that employees already use. When a message contains sensitive information, the employee chooses Protect and Send, and the sensitive attachments and supported message content are encrypted and removed from the outgoing email, which carries a secure link in their place, while the subject line remains standard email text. Recipients open the content through that link without creating an account or installing software, so employees share the way they already do, and the recipients face no new barrier to doing their part.
Once the email has gone, the sender can see when protected content is accessed and can revoke future access, whether the message reached the wrong person or the work has simply finished. Revocation cannot reverse anything a recipient has already viewed or copied, but it ends further access from that point on, which gives the organization a way to limit further exposure from a misdirected send. The content stays protected at rest, in motion, and in use, so control over sensitive information continues after the send instead of ending at it.
Confidencial for Outlook is designed to complement Microsoft Purview and existing DLP investments, not replace them. Those tools continue to identify and govern sensitive information, and Confidencial for Outlook gives employees a protected way to share it when the work requires, so the organization no longer has to choose between warning, blocking, and letting information go.
How does Confidencial for Outlook work?
Confidencial for Outlook is an add-in that lets employees choose Protect and Send when writing a message. Sensitive attachments and supported message content are encrypted and replaced with a secure link. Recipients open the content without an account or special software, and senders can track access and revoke future access after the email is sent.
Keep control after Send
Sensitive information will keep leaving the organization because the business depends on it, so the decision in front of security leaders is whether control ends at the send or continues with the content, and the next step is to see how Confidencial for Outlook protects it after Send.
