# Overeducated, Ph.D.

> Writings, Regrets, and Re-skillings during the AI Revolution


## Posts

- [Security+ 801 Countdown, T-20: SSE Replaces SASE as Secure Access Is Rebuilt](https://benlamb.net/security-plus-801-countdown-t-20-secure-access-sse-e2ee-oob/): 801 3.2 drops SASE and SD-WAN for Security Service Edge and adds end-to-end encrypted messaging, out-of-band management, and file transfer.
- [Security+ 801 Countdown, T-21: Zero Trust Drops the NIST Plane Vocabulary](https://benlamb.net/security-plus-801-countdown-t-21-zero-trust-principles-to-architecture/): Policy engine, policy administrator, and enforcement point are gone. 801 keeps Zero Trust principles in 1.1 and tests user, device, and app access in 3.2.
- [Security+ 801 Countdown, T-22: OT Replaces ICS and SCADA as Cloud Gets Deployment Models](https://benlamb.net/security-plus-801-countdown-t-22-ot-replaces-ics-cloud-deployment-models/): 801 3.1 folds ICS/SCADA, RTOS, and embedded into OT, demotes IoT to a vector, adds community cloud, and splits its considerations list in two.
- [Security+ 801 Countdown, T-23: New Objective 2.6 Covers the Risks of Using AI](https://benlamb.net/security-plus-801-countdown-t-23-ai-risks-hallucinations-data-loss-agents/): Part 2 of 801's AI objective: data loss, privacy, hallucinations, bias, explainability, ethics, and what session hijacking and code execution mean for agents.
- [Security+ 801 Countdown, T-24: New Objective 2.6 Covers Attacks on AI Models](https://benlamb.net/security-plus-801-countdown-t-24-ai-threats-prompt-injection-poisoning/): 801 adds an AI threats objective. Part 1 covers attacks on models: prompt injection, jailbreaking, poisoning, model manipulation, and evasion.
- [Security+ 801 Countdown, T-25: Indicators of Compromise Now Name Hashes, Timestamps, and Log Tampering](https://benlamb.net/security-plus-801-countdown-t-25-indicators-of-compromise-rebuilt/): 801 renames Indicators to Indicators of compromise and adds hashes, IPs, domains, processes, file artifacts, timestamps, and plaintext strings.
- [Security+ 801 Countdown, T-26: Network and Physical Attacks, With Sniffing In and Bollards Out](https://benlamb.net/security-plus-801-countdown-t-26-network-physical-attacks-rebuilt/): 801 2.5 adds sniffing, spoofing, rogue, and cache poisoning, drops collision and birthday attacks, and swaps Domain 1's bollards for tailgating and skimming.
- [Security+ 801 Countdown, T-27: Social Engineering Gets Whaling, Quishing, and Deepfakes](https://benlamb.net/security-plus-801-countdown-t-27-social-engineering-whaling-quishing-deepfakes/): Phishing, vishing, and smishing move into 801's scenario objective with whaling, spear phishing, quishing, and deepfakes. Pretexting and typosquatting are gone.
- [Security+ 801 Countdown, T-28: Sessions, Identity Providers, and MFA Bypass Join the Attack Surface](https://benlamb.net/security-plus-801-countdown-t-28-identity-sessions-idp-mfa-bypass/): 801 names cookies, session tokens, identity providers, and stale credentials as targets and adds user enumeration, replay, and MFA bypass to credential attacks.
- [Security+ 801 Countdown, T-29: Objective 2.4 Rebuilt Around Attack Surfaces, Secrets, and Public Buckets](https://benlamb.net/security-plus-801-countdown-t-29-attack-surface-inventory-secrets-buckets/): 801 2.4 drops SQLi, XSS, and VM escape for exposures: unpatched, obsolete, and rogue systems, hardcoded secrets, public repositories, and open buckets.
- [Security+ 801 Countdown, T-30: Living off the Land and Fileless Malware Arrive](https://benlamb.net/security-plus-801-countdown-t-30-living-off-the-land-fileless-malware/): 801 adds endpoint-based vectors, including built-in and living-off-the-land tools, plus fileless malware as an attack. Detection shifts from files to behavior.
- [Security+ 801 Countdown, T-31: Threat Vectors Rebuilt for RCS, QR Codes, and Fake CAPTCHAs](https://benlamb.net/security-plus-801-countdown-t-31-threat-vectors-rcs-qr-captcha-macros/): 801 2.3 names the channels attackers use now: RCS and collaboration tools, QR codes, CAPTCHA lures, macros, RTF and PDF files, malicious USB, and edge gear.
- [Security+ 801 Countdown, T-32: Threat Actors Rewritten With Terrorists, Competitors, and Accidents](https://benlamb.net/security-plus-801-countdown-t-32-threat-actors-motivations-rewritten/): 801 2.2 adds terrorist, competitor, and accidental actors, renames nation-state, moves Shadow IT out, and adds influence, notoriety, and extortion motives.
- [Security+ 801 Countdown, T-33: CVSS, CVE, and Threat Intelligence Move Into New Objective 2.1](https://benlamb.net/security-plus-801-countdown-t-33-threat-vulnerability-characteristics-cvss-cve/): Scoring and intelligence leave vulnerability management for a new 2.1. CVE finally gets its correct name, and OSINT and the dark web are no longer listed.
- [Security+ 801 Countdown, T-34: Cryptography Adds Protocols and Stops Naming TPM, HSM, and Key Stretching](https://benlamb.net/security-plus-801-countdown-t-34-crypto-protocols-in-tools-out/): 801 1.3 adds Protocols under Encryption but drops TPM, HSM, key stretching, blockchain, and every cipher acronym. What you still need to know, and why.
- [Security+ 801 Countdown, T-35: Change Management Becomes a Scenario Objective](https://benlamb.net/security-plus-801-countdown-t-35-change-management-cab-fail-forward/): Domain 1's only scenario objective is now change management. 801 names the change advisory board and adds fail forward beside backout planning.
- [Security+ 801 Countdown, T-36: Objective 1.1 Now Starts With Defense in Depth](https://benlamb.net/security-plus-801-countdown-t-36-defense-in-depth-least-privilege-controls/): 801 merges 701's controls and fundamentals objectives into one 1.1 built on defense in depth, promotes least privilege, and gives categories dual names.
- [Security+ 801 Countdown, T-37: Same Objective Numbers, Different Topics](https://benlamb.net/security-plus-801-countdown-t-37-objective-crosswalk-numbers-and-verbs/): Eleven objective numbers now point at different topics, and scenario objectives rise from 7 to 9. Here is the full 701-to-801 crosswalk, verbs included.
- [Security+ 801 Countdown, T-38: The Blueprint Was Reweighted and Trimmed to 27 Objectives](https://benlamb.net/security-plus-801-countdown-t-38-new-blueprint-domains-weights-objectives/): SY0-801 keeps 90 questions in 90 minutes but reweights all five domains, renames Domain 2, trims to 27 objectives, and prints a 750 passing score.
- [Network Concepts, No Textbook 9: CIDR and Prefix Length](https://benlamb.net/network-concepts-no-textbook-9-cidr-and-prefix-length/): Slash notation counts street bits; /24 is the same line as mask 255.255.255.0.
- [Network Concepts, No Textbook 8: Same Street? The Subnet Mask](https://benlamb.net/network-concepts-no-textbook-8-same-street-the-subnet-mask/): The subnet mask draws the line between street and house so a host can decide same network or gateway.
- [Network Concepts, No Textbook 7: What an IP Address Is](https://benlamb.net/network-concepts-no-textbook-7-what-an-ip-address-is/): A MAC address names the card; an IPv4 address names where that card sits on the network.
- [Network Concepts, No Textbook 6: Twisted Pair and Category Ratings](https://benlamb.net/network-concepts-no-textbook-6-twisted-pair-and-category-ratings/): The twist fights crosstalk; Cat 5e, 6A, and 8 are the three grades worth remembering.
- [Network Concepts, No Textbook 5: Copper, Fiber, or Radio](https://benlamb.net/network-concepts-no-textbook-5-copper-fiber-or-radio/): Every network link on earth is copper, fiber, or radio; there is no fourth medium.
- [Network Concepts, No Textbook 4: Cable Management Is Documentation You Can Touch](https://benlamb.net/network-concepts-no-textbook-4-cable-management-is-documentation/): Neat cabling is not vanity. A dressed rack is a diagram that never goes stale.
- [Network Concepts, No Textbook 3: Label Both Ends](https://benlamb.net/network-concepts-no-textbook-3-label-both-ends/): A cable labeled on one end is a cable labeled on neither. Labeling is the cheapest documentation a network will ever get.
- [Network Concepts, No Textbook 2: Network Scopes](https://benlamb.net/network-concepts-no-textbook-2-network-scopes/): LAN, WAN, and their cousins name distance and ownership, not fashion labels for networks.
- [Network Concepts, No Textbook 1: Why We Skip OSI Early](https://benlamb.net/network-concepts-no-textbook-1-why-we-skip-osi-early/): ITN 101 postpones the OSI model until students have cables, boxes, and addresses to hang the seven floors on.
- [AI Power Tips 5: Iterate, Disclose, and Explain It Back](https://benlamb.net/ai-power-tips-5-iterate-disclose-and-explain-it-back/): Improve the draft against the assignment, say how you used AI, and be ready to explain every part of what you submit.
- [AI Power Tips 4: Four Questions Before You Trust It](https://benlamb.net/ai-power-tips-4-four-questions-before-you-trust-it/): Assumptions, logic, evidence, and consequences: a short routine for catching AI mistakes before they become yours.
- [AI Power Tips 3: Delegate Only What Is Permitted](https://benlamb.net/ai-power-tips-3-delegate-only-what-is-permitted/): A Delegation Map sorts each assignment into what AI may do, what it may assist with, and what you must do alone.
- [AI Power Tips 2: Know Which Model Is Answering](https://benlamb.net/ai-power-tips-2-know-which-model-is-answering/): Fast models and reasoning models are different tools. Check which one is active and match it to the task.
- [AI Power Tips 1: Using AI Well Is a Study Skill](https://benlamb.net/ai-power-tips-1-using-ai-well-is-a-study-skill/): Most students already use AI. The gaps are in how well they use it and who can pay for the better tools.
- [Neurodivergent Friendly Teaching 11: Telling Students, or Not](https://benlamb.net/neurodivergent-friendly-teaching-11-telling-students-or-not/): Disclosure is your call, and describing a practice is not the same as naming an identity.
- [Neurodivergent Friendly Teaching 10: Access Is Not Only for Students](https://benlamb.net/neurodivergent-friendly-teaching-10-access-is-not-only-for-students/): Naming the workplace barriers that make teaching harder, and asking your college to help remove them.
- [Neurodivergent Friendly Teaching 9: Energy Is a Budget](https://benlamb.net/neurodivergent-friendly-teaching-9-energy-is-a-budget/): Quiet transitions, reusable templates, and a lighter lesson for the days when the full one is not available.
- [Neurodivergent Friendly Teaching 8: One Question Channel and a Planned Pause](https://benlamb.net/neurodivergent-friendly-teaching-8-one-question-channel-and-a-planned-pause/): Separating demonstration from troubleshooting, and telling students when to expect an answer.
- [Neurodivergent Friendly Teaching 7: The Cue Sheet, or How I Stopped Losing Thursday](https://benlamb.net/neurodivergent-friendly-teaching-7-the-cue-sheet-or-how-i-stopped-losing-thursday/): Externalize the live sequence so working memory can do something better than remember what comes next.
- [Neurodivergent Friendly Teaching 6: Grade the Skill, Not the Format](https://benlamb.net/neurodivergent-friendly-teaching-6-grade-the-skill-not-the-format/): What a timed test measures, when to offer an alternative format, and when to hold the line.
- [Neurodivergent Friendly Teaching 5: Noise, Light, and the Two-Minute Warning](https://benlamb.net/neurodivergent-friendly-teaching-5-noise-light-and-the-two-minute-warning/): Small moves on sensory load and transitions, and why you should ask before scheduling a break.
- [Neurodivergent Friendly Teaching 4: Participation Is Not a Speed Contest](https://benlamb.net/neurodivergent-friendly-teaching-4-participation-is-not-a-speed-contest/): Wait time, writing first, and group roles that let quiet students show what they know.
- [Neurodivergent Friendly Teaching 3: One Thing at a Time in the Lab](https://benlamb.net/neurodivergent-friendly-teaching-3-one-thing-at-a-time-in-the-lab/): Software demonstrations ask students to do four things at once. Here is how to ask for one.
- [Neurodivergent Friendly Teaching 2: Be Explicit](https://benlamb.net/neurodivergent-friendly-teaching-2-be-explicit/): Five things every assignment should state, and the one place they should live.
- [Neurodivergent Friendly Teaching: Introduction](https://benlamb.net/neurodivergent-friendly-teaching-introduction/): Working definitions, a premise, and a map of the ten posts to come.
- [ChatGPT at Work 11: How Not to Get Fired Using AI](https://benlamb.net/chatgpt-at-work-11-how-not-to-get-fired-using-ai/): Workplace AI requires approved tools, careful data handling, verified outputs, and clear limits on what the system may do.
- [ChatGPT at Work 10: Your Files Are Part of the Prompt](https://benlamb.net/chatgpt-at-work-10-your-files-are-part-of-the-prompt/): Your project files supply context the chat prompt does not have to repeat. Prepare the folder, identify authoritative sources, and check the requirements.
- [ChatGPT at Work 9: GPT-Live Can Listen While It Talks](https://benlamb.net/chatgpt-at-work-9-gpt-live-can-listen-while-it-talks/): GPT-Live can listen while it speaks, making AI conversation more natural. A classroom interview explores the promise and the need to check its claims.
- [ChatGPT at Work 8: To Chat or To Work? That Is The Question](https://benlamb.net/chatgpt-at-work-8-to-chat-or-to-work/): Use Chat to think through a problem and Work to delegate a deliverable. The difference involves the tools and process around the model.
- [ChatGPT at Work 7: Editable Writing Blocks](https://benlamb.net/chatgpt-at-work-7-editable-writing-blocks/): The newer inline writing block makes collaborative editing faster, more natural, and far less clunky than the older Canvas interface.
- [ChatGPT at Work 6: The First Response Is Slop](https://benlamb.net/chatgpt-at-work-6-the-first-response-is-a-draft/): Useful iteration means blunt feedback, multiple attempts, strategic restarts, and knowing when to try another model—or stop.
- [ChatGPT at Work 5: Verify Before You Act](https://benlamb.net/chatgpt-at-work-5-verify-before-you-act/): ChatGPT can assist with consequential IT work, but people must still verify evidence, protect data, and approve the action.
- [ChatGPT at Work 4: Turn Good Prompts Into Workflows](https://benlamb.net/chatgpt-at-work-4-turn-good-prompts-into-workflows/): The value of ChatGPT grows when a successful prompt becomes a documented, measurable workflow another person can run.
- [ChatGPT at Work 3: Start With the Business Question](https://benlamb.net/chatgpt-at-work-3-start-with-the-business-question/): Useful AI analysis starts with a business question, clean data, and a deliverable someone can act on.
- [ChatGPT at Work 2: Make It Show Its Sources](https://benlamb.net/chatgpt-at-work-2-make-it-show-its-sources/): Business research with ChatGPT depends on current sources, visible limitations, and a clear line from evidence to decision.
- [ChatGPT at Work 1: Define the Job](https://benlamb.net/chatgpt-at-work-1-define-the-job/): A useful ChatGPT prompt defines the job, context, constraints, and form of the finished work before generation begins.
- [Before and After Networks 10: The Telephone, From Circuits to Packets](https://benlamb.net/before-after-networks-10-telephone/): The phone call's quiet revolution: from a reserved circuit to best-effort packets, with dates.
- [Before and After Networks 9: When Things Break](https://benlamb.net/before-after-networks-9-when-things-break/): Isolation used to contain failure. Connection propagates it.
- [Before and After Networks 8: The Pace of Life](https://benlamb.net/before-after-networks-8-pace-of-life/): We engineered the waiting out of our systems — and some of the patience out of ourselves.
- [Before and After Networks 7: Getting Around](https://benlamb.net/before-after-networks-7-getting-around/): A paper map is a snapshot. A navigation app is a live feed.
- [Before and After Networks 6: Entertainment](https://benlamb.net/before-after-networks-6-entertainment/): Entertainment ran on schedules and shelves until networks dissolved both.
- [Before and After Networks 5: Shopping and Banking](https://benlamb.net/before-after-networks-5-shopping-banking/): Trust moved from marble lobbies to protocols, and the attack surface moved with it.
- [Before and After Networks 4: Work and School](https://benlamb.net/before-after-networks-4-work-school/): When work became information flow, it went wherever packets go.
- [Before and After Networks 3: Communities](https://benlamb.net/before-after-networks-3-communities/): Community used to mean proximity. Networks broke that link — for better and worse.
- [Before and After Networks 2: Knowledge](https://benlamb.net/before-after-networks-2-knowledge/): Finding information used to be an errand. Now the hard part is deciding what to believe.
- [Before and After Networks 1: Messages](https://benlamb.net/before-after-networks-1-messages/): Networks made message delivery nearly free—and attention, privacy, and security the new bottlenecks.
- [The Knot in the Network](https://benlamb.net/the-knot-in-the-network/): Why the word node has always described a position in a network rather than a particular kind of device.
- [Four Rules for Governing World Models](https://benlamb.net/four-rules-for-governing-world-models/): Independent testing, worker authority, engineered limits, and clear liability should come before world-model robots enter workplaces.
- [The Policy Problem of Physical AI](https://benlamb.net/policy-problem-of-physical-ai/): Physical AI moves experimental prediction into industrial control before workers or regulators have reliable ways to judge whether it is safe.
- [When a World Model Makes the Wrong Physical Decision](https://benlamb.net/when-world-model-makes-wrong-physical-decision/): When world models control machines, errors can injure workers, erase practical knowledge, eliminate jobs, and diffuse responsibility.
- [A World Model Enters the Factory](https://benlamb.net/world-model-enters-factory/): FLUX-mimic is being tested on Audi assembly lines, bringing generative-video world models into robot control and real production work.
- [Plausible Physics Is Not Physical Understanding](https://benlamb.net/plausible-physics-is-not-physical-understanding/): A world model can produce a future that looks plausible without understanding the physical rules that make it possible—or safe.
- [What Is a World Model?](https://benlamb.net/what-is-a-world-model/): World-model research joins two ambitions: representing the present and predicting the future. The shared label can conceal the gap.
- [World Models: The Next Frontier of AI Hype](https://benlamb.net/world-models-next-frontier-ai-hype/): World models promise AI that can predict and act in physical environments. Their impact may be immense, but the safety case remains speculative.
- [Why Genealogy?](https://benlamb.net/why-genealogy/): Genealogy is the history of things that claim to have no history—and a method for unsettling what the present treats as natural or necessary.
- [Digital Literacy From Above and Below](https://benlamb.net/digital-literacy-from-above-and-below/): How Brian Street’s models reveal the difference between institutionally sanctioned digital literacy and vernacular literacies.
- [Operate, Evaluate, Critique](https://benlamb.net/operate-evaluate-critique/): How computer, information, and media literacy became the operate–evaluate–critique structure inside digital and AI literacy.
- [The Two Pauls of Contemporary Literacy Claims](https://benlamb.net/two-pauls-of-contemporary-literacy-claims/): How Paul Zurkowski’s information literacy became Paul Gilster’s digital literacy—and why their shared problem persists in the AI age.
- [Toward a Genealogy of Digital Literacy](https://benlamb.net/toward-a-genealogy-of-digital-literacy/): How literacy became a measurable, governable category—and how digital literacy inherited its promises, exclusions, and demand for re-skilling.
- [Digital Literacy Is a Moving Target](https://benlamb.net/digital-literacy-is-a-moving-target/): How digital literacy became a social minimum—and why AI literacy extends the demand that workers continually adapt to new technological baselines.
- [Digital Literacy Has Three Sides](https://benlamb.net/digital-literacy-has-three-threads/): My working definition of digital literacy: using technology effectively, efficiently, and responsibly.
- [Digital Literacy Is a Terrible Name for a Useful Class](https://benlamb.net/digital-literacy-is-a-terrible-name-for-a-useful-class/): Why hands-on office software skills may be more valuable than the corny course name suggests.
- [I Am Overeducated. That Is Not the Same as Smart.](https://benlamb.net/i-am-overeducated-that-is-not-the-same-as-smart/): A lifelong student reflects on the difference between credentials, intelligence, useful skills, and a life well spent.
- [Hello, blog!](https://benlamb.net/hello-blog/): The obligatory first post — what this blog is, who's writing it, and what the three words in the subtitle actually mean.

## Pages

- [Contact](https://benlamb.net/contact/): Write to Professor Lamb, or subscribe by RSS or email digest.
- [Blog](https://benlamb.net/blog/): The obligatory first post — what this blog is, who's writing it, and what the three words in the subtitle actually mean.
- [About](https://benlamb.net/about/): Professor Lamb teaches IT at a community college. Overeducated, Ph.D. is writings, regrets, and re-skillings during the AI revolution.
- [Fresh posts](https://benlamb.net/): 801 3.2 drops SASE and SD-WAN for Security Service Edge and adds end-to-end encrypted messaging, out-of-band management, and file transfer.
- [Policy](https://benlamb.net/policy/)
- [Privacy Policy](https://benlamb.net/privacy-policy/): Our website address is: http://benlamb.net.

## Optional

- [Agent (MCP protocol)](websites-agents.hostinger.com/benlamb.net/mcp)

[comment]: # (Generated by Hostinger Tools Plugin)