Risks of Moving Data to the Cloud, Explained

UPI Study › Courses › Ethics in Technology › Free lessons › Lesson 8.6 Free with a free account

Risks of Moving Data to the Cloud

Ethics in Technology · Chapter 8: Guarding the Data Vault: Database Protection, Cloud Risks, and Intrusion DetectionSection 8.6 · lesson 40 of 11615 min · free with a free account

← The Cloud Security ModelNext lesson: Intrusion Detection Systems →

Earn 3 college credits for Ethics in TechnologyRecommended for credit by NCCRS · Computer Science 195 · self-paced · one payment, lifetime access
Enroll in Ethics in Technology · $250 →Download the syllabus (PDF)

In this lesson

Configuration, APIs, identities, shared infrastructure, recovery, and dependency.

  • Explain continuing duties and common cloud misconfigurations
  • Evaluate API authorization, stolen identities, and tenant isolation
  • Distinguish durability, recovery, and provider portability
  • Analyze the Capital One case and the ethical duties behind its controls

Free study notes

Get this lesson's notes by email

The key points and 4 practice questions with answers, in your inbox. One email, free.

One email. Unsubscribe in one click.

Transcript

Everything said in the video, section by section (1,801 words). Read along, or search it.

Chapter 8

Risks of Moving Data to the Cloud

Moving information to the cloud changes operating duties and the paths through which data can be exposed. Provider expertise and maintained infrastructure can help, but overall security still depends on the actual arrangement. This lesson begins with continuing customer decisions about identities, access, configuration, data use, and recovery. We examine common configuration errors and structural controls that make safe choices easier. Programming interfaces need both authentication and authority for each applicable action and record. Stolen identities can bypass a password-only boundary, as the source's historical Snowflake example illustrates. Shared infrastructure introduces isolation, availability, and other risks that deserve a careful assessment. Durability, historical recovery, and independence protect against different loss routes. Portability also depends on the services and operating assumptions built around the data. We then trace the reported Capital One incident, connecting its boundaries with defensive controls. Finally, we consider ethical accountability and a hospital's choice between shared and dedicated arrangements.

Continuing Duties

Managed Infrastructure Still Needs Safe Customer Use

A managed service transfers specified operating duties, rather than every security decision or obligation. Check the actual service boundary and the controls available to the customer. Storage readable by everyone, or an unexpectedly broad signed-in population, can expose records. Database and management interfaces may also be reachable beyond the boundary their work requires. Excessive roles let people or applications perform more actions than their tasks justify. Relevant logs may be missing, incomplete, or present without meaningful review. Encryption and recovery settings sometimes go unchecked because someone assumes the defaults are sufficient. Forgotten tests, temporary services, and obsolete accounts can remain active and exposed. Customers need owners for their applicable data, identities, configuration, integrations, and recovery decisions. Provider safeguards, contracted work, and applicable legal duties continue around those customer decisions. The practical task is to resolve the boundary and verify the controls, without leaving an unowned gap.

Configuration Discipline

Fast Provisioning Needs Review, Guardrails, and Evidence of the Running State

Cloud resources can be provisioned quickly, removing informal delays that once created opportunities for review. Many services and settings also make a complete understanding difficult for any one person. Some resources have broad remote reach, although cloud deployments can also use restricted boundaries. Infrastructure as code makes intended configuration reviewable and versioned. Its files do not prove that deployed state and external dependencies will always remain identical. Secure baselines and resource inventories help teams understand the environment they actually operate. Scoped guardrails can prevent selected unsafe choices, subject to coverage, exceptions, and bypass privileges. Continuous checks help identify unsafe settings and drift from intended configuration. Someone still needs to assess the findings and act in time. Staff training complements these structural controls by explaining the settings and their consequences. Misconfiguration is serious, but this lesson does not invent a universal ranking or instant exposure measurement.

Programming Interfaces

Authentication Does Not Grant Access to Every Object

Programming interfaces connect cloud management and application work to permissions and sensitive operations. The control plane provides interfaces for creating, changing, or deleting infrastructure. Application interfaces also expose business actions and records, requiring appropriate authority checks. Authentication establishes an identity, but does not grant that identity access to every object. Check whether this caller may perform this action on this particular record. Internal and temporary sensitive endpoints still need appropriate protection, even when they are less visible. Service secrets should be protected, narrowly scoped, and managed throughout their usable lifetime. Some credentials expire; others need active rotation, revocation, or a safer authentication arrangement. Consumption limits and monitoring help identify abnormal volume and reduce resource abuse. They supplement authorization, rather than creating a right to read records slowly. Inventory maintained versions and remove obsolete interfaces before forgotten policies become a hidden access route.

Account Hijacking

The 2024 Snowflake Campaign Shows Why Valid Credentials Need Further Protection

The historical Snowflake example concerns a campaign reported by Mandiant in June, twenty twenty-four. Mandiant and Snowflake had notified about one hundred sixty-five potentially exposed organizations. That wording does not establish one hundred sixty-five confirmed breaches within the report's population. Mandiant's investigated incidents traced access to stolen customer credentials, without evidence of a Snowflake enterprise breach. Affected accounts lacked multifactor authentication; stolen credentials remained usable, and network allowlists were absent. These findings support strong authentication, appropriate network boundaries, credential lifecycle controls, and abnormal-access monitoring. They describe historical settings, rather than a claim about present defaults. Snowflake subsequently introduced a phased rollout of stronger authentication requirements. Applications and automation also have identities, which need narrow permissions and protected authentication. Use appropriate temporary credentials or manage longer-lived secrets, with monitoring and recovery arrangements. Provider defaults and customer configuration both influence exposure; neither removes the other's applicable duties.

Multi-tenancy

Shared Hardware Relies on Maintained Logical Boundaries

Multi-tenancy allows customers to share physical resources while maintaining logically separated accounts, data, and configuration. The diagram places a maintained isolation boundary between those tenants and the shared infrastructure. That boundary is real, but software and configuration must continue to enforce it. An isolation escape can cross tenant boundaries; its scope depends on the particular defect. A shared authentication or other service flaw can also expose more than one customer. Another tenant's load or shared-address reputation may affect availability or reachability. Released storage needs effective sanitization or suitable cryptographic protection before reuse. Investigative action affecting shared infrastructure can potentially affect others, depending on jurisdiction and arrangements. Provider responsibilities dominate operation of the shared boundary, without making every customer response useless. Customers can choose services and isolation options, manage access, and reduce exposed data or incident impact. Dedicated infrastructure changes some risks, but does not automatically correct the customer's configuration or permission mistakes.

Data Loss

Durability Protects Against Some Failures, While Recovery Needs Usable Historical Copies

Durability protects against specified failures, commonly including underlying storage-component loss. It does not prove that every valid earlier business state remains recoverable. Replication may reproduce deletion or corruption, depending on the service and its version protections. Ransomware or destructive credentials can also threaten recovery copies if they share the same authority. Lost keys can make a dependent copy unusable without another key-recovery route or recoverable copy. Account suspension and closure require checking actual access, retention, grace, and recovery terms. Provider or facility failure introduces dependencies beyond a single storage device. Separate relevant recovery accounts, credentials, copies, and failure domains where the risk requires it. Check versioning, deletion protection, supported retention locks, and their actual limits. Test restoration to establish that the required copies, keys, and environment can produce usable records. Replication can contribute to resilience, but alone does not establish a backup against every loss route.

Vendor Lock-in

Exporting Data Does Not Automatically Recreate the Working System

Vendor lock-in occurs when changing provider becomes impractical or too costly for the organization's purposes. The obstacle may include much more than transferring the raw data. Managed-service behavior, identity rules, integration patterns, and operational knowledge can need substantial rebuilding. Application code, containers, standard runtimes, and standard queries may transfer if the target supports their assumptions. Differences in dependencies, dialects, deployment, and assurance evidence still require work. Test full exports in usable formats and verify restoration elsewhere before an urgent exit is needed. Understand actual transfer costs and current contract terms, rather than assuming every provider charges the same fees. Isolate provider-specific dependencies behind interfaces when the benefit justifies the engineering cost. Negotiate exit help, data return, and applicable deletion evidence while terms can still be shaped. Portability is a property of the whole workload, rather than one exported file or container. An organization may deliberately accept deep dependency for useful capability, provided it assesses and records the tradeoff.

Capital One, 2019

The Reported Impact Included Applicants Who Could Not Inspect the Bank’s Controls

Capital One disclosed the incident in twenty nineteen, with approximately one hundred six million affected people. The bank reported about one hundred million in the United States and six million in Canada. Application records extended from two thousand five through early twenty nineteen, alongside specified customer and transaction information. Affected categories included contact details, income, credit information, and sensitive identifiers or linked accounts for subsets. The bank reported that credit-card account numbers and login credentials were not compromised. Those limits do not erase the sensitivity of the records taken. An outside report arrived on July seventeenth; the bank confirmed the incident on July nineteenth. The regulator assessed an eighty-million-dollar penalty in twenty twenty, and customer litigation produced a separate settlement. The settlement amount was one hundred ninety million dollars; the attacker was convicted in twenty twenty-two. The regulator cited inadequate pre-migration risk assessment and delayed correction of deficiencies. Criminal misconduct and organizational control failures are distinct responsibilities, requiring separate evidence and conclusions.

Reported Path, Conceptually

Trace Each Boundary, Then Identify a Control That Reduces That Path

The source presents the reported path as a sequence of connected boundaries. A public-facing component accepted an unsafe request that reached an internal credential service. The service returned credentials associated with the component's role. That role could list and read sensitive storage beyond the component's needed work. Records were listed and copied; the bank learned of the incident through an external report. Harden request handling and maintain the component, then restrict unnecessary internal reach. Use stronger supported metadata protections and protect credential issuance. Apply least privilege to the application's actual job, rather than granting broad storage access by convenience. Monitor unusual reads and outbound activity, and ensure findings lead to timely action. Controls can disrupt this described path or reduce impact, without proving one control prevents every attack. External discovery does not mean no records existed; investigation can use logs that failed to trigger a timely response.

Ethical Accountability

People Bear Data Risks They Often Cannot Evaluate or Influence

The ethical concern extends beyond the technical path to the people exposed by organizational decisions. Applicants may never become customers, yet retained records can still carry serious privacy and financial risks. Those people generally cannot inspect the architecture or evaluate each permission decision. Their limited visibility does not reduce the organization's duty to assess and manage its practices. Customer operating work includes configuration review, role limits, internal reach restrictions, detection, and correction. Providers also shape protective capabilities, defaults, responsibility guidance, and the clarity of product limitations. Leadership must resource the work, assess migration risks, resolve findings, and assign accountable owners. Appropriate collection, use, retention, communication, and applicable remedies connect those controls to affected people's interests. Useful safeguards still require competent implementation, time, and resources. A later product change does not alone establish legal liability or prove an earlier safeguard universally sufficient. The central lesson is accountable risk management, rather than assuming the responsibility diagram settles every moral question.

Hospital Decision and Review

Choose the Arrangement from Evidence, Applicable Duties, and Operating Capability

A hospital should compare the actual shared and dedicated arrangements using evidence and applicable requirements. Assess tenant isolation, provider controls, sensitivity, likely consequences, and resilience dependencies. Dedicated infrastructure may change particular exposures, but it does not cure unsafe customer access decisions. Review the hospital's ability to maintain configuration and require appropriate strong authentication. Machine identities, monitored activity, protected recovery copies, and tested restoration remain relevant in either arrangement. Identify applicable health-data duties, jurisdiction, access arrangements, and contract requirements. Include incident cooperation, exit help, and relevant data-return or deletion evidence. Weigh staffing, cost, and the possibility that essential controls will go unmaintained. The review questions connect transferred duties, configuration structure, stolen identities, defensive boundaries, and this hospital decision. Cloud adoption changes risk, while its overall benefit depends on architecture, operation, and requirements. Document a defensible choice for the actual organization, rather than prescribing one universal answer for every hospital.

About Ethics in Technology at UPI Study

Ethics in Technology (Computer Science 195) is a self-paced online course of 116 video lessons in 15 chapters, from cybersecurity basics, privacy and surveillance to big data, hacking, malware, network defense, intellectual property, cybercrime, encryption, emerging technology and digital citizenship. It carries a NCCRS credit recommendation for 3 semester credits.

Every lesson transcript is free on pages like this one, and the videos are free with a free UPI Study account. Enrolling adds a practice quiz after each lesson, the graded work and final exam that earn the credit, lifetime access, and an official transcript on completion. UPI Study has helped over 130,000 students with college credits. ACE and NCCRS have over 2,100 cooperating colleges that will consider accepting credits; the final decision is with the universities.

Enroll now → Course details Download the syllabus All Computer Science courses

Compare: Cybersecurity vs Ethics in Technology · Which ethics course?

Why take it here

Why students take Ethics in Technology with UPI Study

The right and wrong of the digital worldEthics in Technology looks at the choices technology forces on people and organizations: security and privacy, surveillance and the law, big data, hacking and cyber warfare, malware and breaches, intellectual property, cybercrime, encryption, artificial intelligence and responsible digital citizenship. It suits computer science, information technology and business degrees.
Credit at a fraction of the priceThe same 3 credits can cost $750 or more at a US college. Here it is one payment, self-paced, with lifetime access.
Finish your degree soonerTransfer credits reduce the courses you still owe, so you can graduate earlier or lighten a heavy semester.
Strengthen an applicationA completed college-level course on an official transcript shows admissions and scholarship committees you can do the work.
Study on your own scheduleNo start dates, no deadlines. Watch the lessons, practice, and take the final when you are ready.
130,000+students helped with college creditsACE and NCCRS have over 2,100 cooperating colleges that will consider accepting credits; the final decision is with the universities.

Where the credit goes

Universities that work with NCCRS

Ethics in Technology carries a NCCRS credit recommendation. These universities work with NCCRS; some are UPI Study's signed partners. ACE and NCCRS have over 2,100 cooperating colleges that will consider accepting credits; the final decision is with the universities.

Thomas Edison State UniversityThomas Edison State UniversitySigned partnerUniversity of the PeopleUniversity of the PeopleSigned partnerCharter Oak State CollegeCharter Oak State CollegeSigned partnerStrayer UniversityStrayer UniversitySigned partnerColumbia Southern UniversityColumbia Southern UniversitySigned partnerIndiana TechIndiana TechSigned partnerPenn StatePenn StateWorks with NCCRSSouthern New Hampshire UniversitySouthern New Hampshire UniversityWorks with NCCRSSUNYSUNYWorks with NCCRS +many more cooperating universitiesFind my college →

Find my college → Not listed? Ask Athena in the corner of this page, or book a 15-minute call.

FAQ

Questions students ask us

About Risks of Moving Data to the Cloud

What are common cloud security misconfigurations?
Common cloud misconfigurations include storage readable by anyone, database or management interfaces reachable from the internet, roles with far more permissions than needed, missing or unreviewed logs, unchecked encryption and backup settings, and forgotten test resources or obsolete accounts. Rapid provisioning and complexity make these easy to create. Infrastructure as code, secure baselines, resource inventories, guardrails and continuous configuration checks help, but someone must act on the findings.
What happened in the 2024 Snowflake data breaches?
In June 2024 Mandiant reported a campaign in which attackers used stolen customer credentials to access Snowflake customer accounts, and about 165 potentially exposed organizations were notified. Investigators found no evidence of a breach of Snowflake's own enterprise environment. The affected accounts lacked multifactor authentication, still had valid old credentials and had no network allowlists. Snowflake then began phasing in stronger authentication requirements.
How did the Capital One data breach happen?
In the 2019 Capital One breach, an attacker tricked a public-facing component into making a request to an internal cloud credential service, a server-side request forgery. The returned credentials belonged to a role that could list and read sensitive storage, so records on about 106 million people in the United States and Canada were copied. The bank learned of it from an outside tip and later paid an $80 million regulatory penalty.
What is vendor lock-in in cloud computing?
Vendor lock-in happens when moving away from a cloud provider becomes impractical or too costly. The barrier is often not the raw data but managed-service behavior, identity rules, integrations and staff knowledge that would need rebuilding. To reduce it, test full exports in usable formats, rehearse restoring elsewhere, understand transfer fees and negotiate exit help and deletion evidence early. Some organizations accept deep dependency deliberately for useful features, after documenting the tradeoff.

About the Ethics in Technology course

How many college credits is Ethics in Technology?
Ethics in Technology (Computer Science 195) carries a NCCRS credit recommendation for 3 semester credits. ACE and NCCRS have over 2,100 cooperating colleges that will consider accepting credits; the final decision is with the universities.
How much does Ethics in Technology cost and what is included?
$250, paid once, with lifetime access: all 116 video lessons, a practice quiz after each, the graded work and final exam, and an official transcript when you finish. It is sold on its own and is not part of UPI Study PRO.
Which universities accept UPI Study credits?
UPI Study has signed partner agreements with universities such as Thomas Edison State University, University of the People and Charter Oak State College, and works with universities such as Penn State, SNHU and SUNY. ACE and NCCRS have over 2,100 cooperating colleges that will consider accepting credits; the final decision is with the universities. Use "Find my college" on this page to check yours.
How long does Ethics in Technology take?
It is self-paced with no deadlines. The 116 lessons total about 28 hours of video; most students add reading and practice time and finish in a few weeks to a few months, on their own schedule.
How are the exams done?
Quizzes and assignments are online in the course. The final exam is taken online with proctoring; a person on the UPI Study team checks assignments, usually within two days.
How do my credits reach my college?
When the course shows 100% complete, request the official transcript from your dashboard. It is sent to the college you name, and you can send it to more than one. Check your college first with "Find my college" on this page, or ask Athena.

Ready for the full course? Earn 3 credits in Ethics in Technology.

116 video lessons, a quiz after each one, and the graded work that earns the credit. Recommended for credit by NCCRS.

Enroll in Ethics in Technology →Download the syllabus
Ethics in Technology
3 credits · NCCRSEnroll →