Tygart Media

Tag: Anthropic

  • The Restoration Talent Window Is Closing Faster Than You Think

    The Restoration Talent Window Is Closing Faster Than You Think

    Last refreshed: May 15, 2026

    A LinkedIn post from a restoration recruiter in Houston tipped me off this morning. He’s right — but the timeline is shorter than most people in the industry realize.

    Mitchell Riley LinkedIn post about Claude Managed Agents announcement
    Mitchell Riley’s LinkedIn post that started this train of thought.

    This article is part of The Restoration Operator’s Playbook — Tygart Media’s body of work on how the industry’s best restoration companies are actually thinking in 2026. Start with the pillar piece if this is your first read.

    The post that got me thinking

    This morning I logged into LinkedIn and saw a post from Mitchell Riley — a restoration industry recruiter in Houston who places PMs, GMs, and business development leaders for restoration contractors across the country. Mitchell flagged Anthropic’s Claude Managed Agents launch with the kind of casual enthusiasm only people who actually use this stuff every day can manage. He called it “pretty cool” and noted that Claude will now build you an agent based on natural language.

    He’s right. He’s also pointing at something most of the restoration industry hasn’t fully processed yet.

    What Anthropic actually shipped

    On April 8, 2026, Anthropic launched Claude Managed Agents in public beta. The short version: the infrastructure work that used to take three to six months of engineering — sandboxed code execution, credential management, long-running session persistence, error recovery, observability — is now a managed service. You define what the agent should do. Anthropic runs it.

    The companies already shipping production agents on it: Notion, Asana, Rakuten, and Sentry. Notion lets teams delegate coding, slides, and spreadsheets to Claude without leaving the workspace. Rakuten deployed specialist agents across product, sales, marketing, finance, and HR — each live in under a week. Sentry built an agent that goes from flagged bug to open pull request, fully autonomous.

    Internal Anthropic testing showed up to a 10-point improvement in task success on structured generation work versus a standard prompting loop, with the largest gains on the hardest problems.

    That’s the announcement. Here’s why it matters for restoration.

    The bottleneck just moved

    For the last two years, the question every restoration owner asked about AI was some version of: “Can it actually do the work?” The honest answer was usually “not yet, not without a developer team you don’t have.”

    That’s no longer the question. The infrastructure gap closed on April 8. The new bottleneck is not “can you build the agent” — it’s “do you have the human operators who know what the agent should be doing in the first place.”

    Restoration is an industry where the real intelligence lives in people. A senior PM who has worked five hundred losses knows things that have never been written down anywhere. How a Cat 3 storm response actually sequences when the carrier is dragging on TPA approvals. The difference between a contents pack-out that closes clean and one that becomes a six-month dispute. Which mitigation decisions buy you a profitable job and which ones bury you on the reconstruction side. None of that lives in a textbook. It lives in the heads of people who have been doing the work for fifteen or twenty years.

    That tribal knowledge is now the constraint. The companies that win the next three years will be the ones who pair Managed Agents (or something like it) with senior operators who can tell the agent what good looks like. The companies that try to skip that step — that try to hire generalists and teach them restoration on the fly while their competitors are distilling twenty-year veterans into operational systems — are going to get lapped.

    Buy the talent now

    This is where the recruiting angle gets interesting. Senior restoration talent has always been hard to find. It’s about to get much harder, for a reason most owners haven’t priced in yet: the value of a senior PM is no longer just the work that PM does directly. It’s the work an entire AI system does in their image once their judgment has been encoded into the workflow.

    Right now, that arbitrage is open. The market hasn’t repriced senior operators for what they’re actually worth in an AI-augmented restoration company. In twelve to twenty-four months, it will. The owners who hire the best PMs, GMs, and BD leaders now — and who pair them with someone like Mitchell who actually understands the placement game — are going to look like geniuses in 2027.

    Mitchell is one of the people who gets this from the inside. He uses the AI tools himself. He builds workflows. He analyzes things in dimensions and context that most recruiters never touch — most recruiters in this industry are still working from a spreadsheet of resumes and a cell phone. Mitchell is the kind of recruiter who notices when Anthropic ships something that’s going to change the value of every senior hire he places, and posts about it on a Wednesday morning. That’s the level of operator the smart restoration owners are going to want in their corner.

    What to actually do this quarter

    If you run a restoration company and you read this far, three concrete things:

    One. Identify your two or three most senior operators — the people whose judgment is load-bearing for the business. Start documenting how they think, not just what they do. The documentation is the raw material every future AI workflow will run on.

    Two. Open one or two senior hires you’ve been putting off. The talent market is going to tighten. Get in front of it.

    Three. Stop treating AI as an IT project. It’s an operational capability. The companies that figure this out are not waiting for their tech vendor to sell them an “AI feature.” They’re hiring the operators, capturing the judgment, and pointing the tooling at the result.

    Mitchell’s post was three sentences. The full version of what he was pointing at takes about a thousand words. This is that version.

    If you’re a restoration owner thinking about senior placements in the next two quarters, you should be talking to Mitchell. And if you’re thinking about how to operationalize AI inside your company — distilling senior judgment into systems your whole team can run — that’s the conversation we have at Tygart Media.

    Read next: The New Restoration Operator: How the Industry’s Best Companies Are Thinking in 2026 — the pillar piece this article belongs to.

  • Claude for Education: How the University Program Works and How to Get Access

    Claude for Education: How the University Program Works and How to Get Access

    Last refreshed: June 9, 2026

    Claude AI · Fitted Claude

    Claude for Education is Anthropic’s official program for higher education institutions — a university-wide plan that gives enrolled students, faculty, and staff access to Claude’s premium features, including advanced models, learning mode, and API credits for research. It’s institution-facing, not student-facing: your university signs up, and access flows through your .edu email.

    Access: claude.com/solutions/education — for institutions. If your university is already a partner, sign in to claude.ai with your .edu email and your account will be upgraded automatically.

    What Claude for Education Includes

    Feature What it means for your institution
    Campus-wide access Students, faculty, and staff all covered under one institutional agreement
    Learning mode Claude guides students through problems rather than just giving answers — designed to build understanding, not bypass it
    API credits for research Faculty can access the Claude API to accelerate research — dataset analysis, text processing, building learning tools
    Claude Code access Students in technical programs get Claude Code for pair programming and software development learning
    Training and support Anthropic provides implementation resources and ongoing support for faculty and administrators
    Data compliance Anthropic only uses data for training with explicit permission; security standards meet institutional compliance needs

    How to Get Your Institution Enrolled

    The Claude for Education program is applied for by institutions, not individual students. The process runs through Anthropic’s sales team:

      Before You Talk to Anthropic Sales

      I help teams assess Claude fit and avoid overpaying before they enter a sales process. Free 15-minute call — no pitch.

      Email Will First → will@tygartmedia.com

    1. Visit claude.com/contact-sales/education-plan
    2. Submit your institution’s information and intended use case
    3. Anthropic reviews and negotiates the institutional agreement
    4. Once enrolled, students and staff access Claude by signing in with their .edu email

    If you’re a student or faculty member who wants your institution to join, raise it with your IT department, library services, or educational technology office. Anthropic’s first confirmed design partner is Northeastern University (50,000 students and staff across 13 campuses worldwide), and the partner list has been expanding through 2025 and 2026.

    Learning Mode: What Makes the Education Program Different

    The distinctive feature of Claude for Education is learning mode — Claude’s approach shifts from answering questions to guiding students toward answers. Rather than writing the essay or solving the problem directly, Claude asks clarifying questions, prompts reflection, and helps students develop their own reasoning. Anthropic designed this explicitly to strengthen critical thinking rather than bypass it.

    This is a meaningful distinction from standard Claude Pro: the same powerful model, but oriented toward building understanding rather than delivering outputs. For educators concerned about AI undermining the learning process, learning mode is Anthropic’s answer.

    Claude for Education vs Claude for Research

    Faculty and researchers at accredited institutions who need API access for research projects can also apply for Anthropic’s grant programs independently of the campus-wide Education plan. These grants typically provide API credits for research workloads — analyzing datasets, processing large text corpora, building research tools — rather than subscription discounts. Contact Anthropic through their research or social impact team for grant program information.

    Student Programs Within the Education Ecosystem

    Alongside the institutional program, Anthropic runs student-facing programs that provide individual access:

    • Campus Ambassadors — Selected students receive Pro access and API credits in exchange for leading AI education initiatives on campus. Applications open periodically; watch claude.com/solutions/education for current status.
    • Builder Clubs — Student clubs that organize hackathons and demos receive Pro access and monthly API credits. Open to all majors.

    For a full breakdown of how students can access Claude at reduced cost, see Claude Student Discount: The Truth and Legitimate Ways to Save.

    Frequently Asked Questions

    What is Claude for Education?

    Claude for Education is Anthropic’s institutional program for universities — a campus-wide plan covering students, faculty, and staff with premium Claude access including learning mode, API credits for research, and Claude Code. It’s applied for by institutions through Anthropic’s sales team, not individual students.

    How do I access Claude for Education as a student?

    Sign in to claude.ai with your .edu email. If your institution is an Anthropic education partner, your account will be upgraded automatically. If not, ask your IT department or library about joining the program. Alternatively, apply for the Campus Ambassador program or join a Builder Club if available at your school.

    Is Claude for Education free for students?

    For students at partner institutions, yes — access is free through the institutional agreement. Anthropic and the university negotiate the pricing; it’s not passed on to individual students. For students at non-partner schools, there is no individual student pricing — the standard free and paid plans apply.

    Confirmed Claude for Education Partners

    The Claude for Education program has expanded significantly since launch. Confirmed institutional partners and program collaborations include:

    University-Wide Campus Agreements

    • Northeastern University — Anthropic’s first university design partner, providing access to 50,000 students, faculty, and staff across 13 global campuses. Northeastern is collaborating directly with Anthropic on best practices for AI integration in higher education and frameworks for responsible AI adoption.
    • London School of Economics and Political Science (LSE) — Campus-wide rollout focused on equity of access, ethics, and skills development for students entering an AI-transformed workforce.
    • Champlain College — Vermont-based institution with full campus access for students, faculty, and administrators.

    Multi-Institution Programs

    • CodePath Partnership — Anthropic partnered with CodePath, the nation’s largest provider of collegiate computer science education, to put Claude and Claude Code at the center of CodePath’s curriculum. The partnership reaches more than 20,000 students at community colleges, state schools, and HBCUs. Over 40% of CodePath students come from families earning under $50,000 a year, making this program a meaningful equity initiative. Courses include Foundations of AI Engineering, Applications of AI Engineering, and AI Open-Source Capstone.
    • American Federation of Teachers (AFT) — Anthropic is partnering with AFT to offer free AI training to AFT’s 1.8 million members across the United States.
    • Internet2 — Anthropic joined the Internet2 community and is participating in a NET+ service evaluation, working toward broader integration with research and education networks.
    • Instructure — Partnership to embed Claude into Canvas LMS, Instructure’s learning management system used by thousands of institutions.

    International Education Initiatives

    • Iceland — One of the world’s first national AI education pilots, launched with the Icelandic Ministry of Education and Children, providing teachers across the country access to Claude.
    • Rwanda — Partnership with the Rwandan government and ALX bringing a Claude-powered learning companion to hundreds of thousands of students and young professionals across Africa.

    U.S. Federal Commitment

    Anthropic signed the White House’s “Pledge to America’s Youth: Investing in AI Education,” committing to expand AI education nationwide through investments in cybersecurity education, the Presidential AI Challenge, and a free AI curriculum for educators.

    If your institution isn’t on this list, the program is actively expanding — application is through Anthropic’s education team at claude.com/contact-sales/education-plan.

    Claude for Education vs ChatGPT Edu

    Anthropic’s Claude for Education and OpenAI’s ChatGPT Edu are the two major institutional AI offerings competing for higher education partnerships. Both provide campus-wide access at negotiated institutional rates rather than individual student pricing. Here’s how they compare:

    Feature Claude for Education ChatGPT Edu
    Launched April 2025 May 2024
    Pedagogical approach Learning Mode — guides reasoning rather than providing answers directly Standard ChatGPT interface with educator controls
    First design partner Northeastern University University of Pennsylvania (Wharton)
    Notable partners Northeastern, LSE, Champlain, CodePath (20,000+ students) Columbia, Wharton, Oxford, California State University system
    Data privacy default Conversations not used for model training without explicit permission Enterprise-grade privacy with admin controls
    LMS integration Canvas (via Instructure partnership) Multiple LMS integrations available
    Pricing Negotiated per institution; not publicly disclosed Negotiated per institution; not publicly disclosed

    The most distinctive difference is pedagogical philosophy. Claude’s Learning Mode is purpose-built around guided reasoning — Claude is designed to ask questions, prompt students to think through problems, and develop critical thinking rather than provide direct answers. ChatGPT Edu provides the standard ChatGPT experience with administrative controls layered on top.

    For institutions deciding between the two, the real evaluation criteria are usually: which model performs best for your dominant use cases (Claude tends to lead on writing, analysis, and reasoning; ChatGPT often leads on multimodal generation), which integrates better with your existing LMS, and which vendor’s pricing and contract terms work for your procurement process.

    What Claude for Education Actually Costs

    Anthropic does not publish standard pricing for Claude for Education. The program is sold as institutional agreements negotiated between Anthropic’s education team and the school. The factors that drive pricing typically include:

    • Number of users — students, faculty, and staff who will receive access
    • Scope of access — which Claude features, models, and tools are included
    • API credit allocation — for faculty research and student builder projects
    • Contract length — multi-year commitments often produce better per-user economics
    • Compliance and integration requirements — SSO, SCIM, Canvas integration, and other institutional infrastructure

    For institutions sizing their budget before formal conversations, the practical reference point is what Anthropic charges enterprise customers. Anthropic’s Enterprise plan provides per-seat pricing in a similar institutional structure — though education program pricing is typically more favorable than commercial Enterprise rates given Anthropic’s strategic interest in academic adoption.

    The fastest way to get accurate pricing for your institution is to contact Anthropic’s education team at claude.com/contact-sales/education-plan with your user count and use case priorities.

    Building the Case for Your University to Adopt Claude for Education

    If you’re a faculty member, IT administrator, or student trying to get your institution to adopt Claude for Education, the following points have been most effective in conversations with academic procurement teams:

    Pedagogical Alignment

    Claude’s Learning Mode is purpose-built around guided reasoning rather than answer-delivery. This addresses one of the most common faculty objections to AI in education: that students will use AI to bypass learning rather than enhance it. Learning Mode is the structural answer — Claude is designed to prompt students to think rather than think for them.

    Privacy and Compliance

    Anthropic provides explicit assurance that student and faculty conversations are not used for model training without permission. Security standards meet the compliance requirements typical of higher education procurement, including data residency considerations and audit controls. For institutions with FERPA requirements, the Education program is structured to support compliant deployment.

    Equity of Access

    Campus-wide access through institutional agreement removes the financial barrier that exists when AI tools are accessed by individual paid subscriptions. Students from lower-income backgrounds get the same access as students who could otherwise afford a $20/month Pro plan — eliminating an emerging form of academic inequality.

    Research Capability

    Faculty and graduate researchers gain access to API credits and the 1M token context window for processing large datasets, conducting literature reviews, analyzing research corpora, and building research tools. This is meaningful capability that would otherwise require individual API budgets.

    Integration with Existing Infrastructure

    The Instructure partnership for Canvas LMS integration and the Internet2 NET+ service evaluation reduce the integration burden on institutional IT teams. Claude for Education is designed to plug into the existing edtech stack rather than require a parallel system.

    Practical Next Steps for Internal Advocates

    1. Document specific use cases at your institution — what would students, faculty, and administrators actually do with Claude
    2. Identify a faculty champion or department head willing to sponsor a pilot
    3. Connect with your institution’s IT or educational technology office to understand procurement requirements
    4. Have your institutional leadership contact Anthropic at claude.com/contact-sales/education-plan for a formal evaluation conversation

    Claude for K-12 and Teacher Training

    While Claude for Education is primarily focused on higher education institutions, Anthropic has expanded into K-12 and teacher development through several pathways:

    • American Federation of Teachers partnership — Free AI training for AFT’s 1.8 million teacher members. This is one of the largest teacher AI training initiatives in the U.S.
    • Iceland national pilot — National-scale AI education pilot with the Icelandic Ministry of Education and Children, providing classroom teachers across the country access to Claude. This is one of the world’s first national-scale AI education programs.
    • White House Pledge to America’s Youth — Anthropic’s commitment to expand AI education through cybersecurity education investments, the Presidential AI Challenge, and free AI curriculum for educators.

    For K-12 schools and individual teachers wanting to bring Claude into the classroom, the formal Education program is currently structured around higher education. K-12 institutions interested in formal partnerships should still reach out via the Education contact channel — Anthropic has been expanding into K-12 through targeted pilots and may have programs available depending on the school’s profile.

    Additional Frequently Asked Questions

    Which universities have Claude for Education access?

    Confirmed campus-wide partners include Northeastern University, the London School of Economics and Political Science, and Champlain College. The CodePath partnership extends Claude access to more than 20,000 students at community colleges, state schools, and HBCUs across the U.S. Internationally, Iceland and Rwanda have national-scale education partnerships. The partner list is actively expanding.

    How is Claude for Education different from Claude Pro?

    Claude Pro is an individual paid subscription at $20/month. Claude for Education is an institutional agreement that provides equivalent access (and often more, including API credits and Learning Mode) to all students, faculty, and staff at participating institutions. Education access is funded by the institution rather than the individual student.

    Does Claude for Education include Claude Code?

    Claude Code access depends on the specific institutional agreement. The CodePath partnership specifically integrates Claude Code into the curriculum, indicating that Claude Code is available within Education program agreements when negotiated. Institutions should confirm Claude Code inclusion as part of their procurement conversation.

    How long does the Claude for Education evaluation process take?

    The timeline varies by institution. Initial conversation through formal contract typically takes weeks to months depending on the institution’s procurement process, security review requirements, and contract complexity. Anthropic’s education team can provide a more specific timeline based on your institutional requirements.

    Can community colleges and smaller institutions join Claude for Education?

    Yes. The CodePath partnership specifically reaches community colleges and HBCUs, and the program is not limited to large research universities. Smaller institutions interested in the program should reach out through the same education contact channel — Anthropic’s expansion strategy is actively focused on reaching institutions that have historically been overlooked in technology partnerships.

    What happens to my Claude for Education access when I graduate or leave the institution?

    Access is tied to your institutional affiliation. When you’re no longer enrolled or employed at the partner institution, your account reverts to the standard Free or Pro tier (depending on whether you choose to subscribe individually). Conversations and Projects you created during your education access typically remain in your account, but premium features will require an individual subscription to continue using.

    Is there a Claude for Education program for graduate students and postdocs specifically?

    Graduate students and postdoctoral researchers at partner institutions are covered under the same campus-wide agreement as undergraduate students. For research-specific API credits at scale, faculty and researchers can also apply for Anthropic’s research grant programs independently of the campus-wide Education plan — these typically provide API credits for research workloads rather than subscription discounts.

    How does Learning Mode actually work?

    Learning Mode shifts Claude’s default response pattern from answer-delivery to guided reasoning. Instead of producing a complete solution to a problem, Claude asks clarifying questions, prompts the student to identify the next step, validates correct reasoning, and surfaces gaps in understanding. The mode is designed to support the educational goal of building student capability rather than completing assignments. Faculty can configure Learning Mode behavior at the institutional level.

    Can faculty use Claude for Education for research that isn’t tied to teaching?

    Yes. The program is designed to support faculty research activity in addition to classroom teaching. API credits within the institutional agreement can be allocated to faculty research projects, including data analysis, literature synthesis, research tool development, and large-scale text processing. The 1M token context window on Opus 4.8 and Sonnet 4.6 makes the program particularly useful for research workflows requiring large context.

  • Claude Jailbreak: How It Works, Why It’s Hard, and What Happens When It Succeeds

    Claude Jailbreak: How It Works, Why It’s Hard, and What Happens When It Succeeds

    Last refreshed: May 15, 2026

    Model Accuracy Note — Updated May 2026

    Current flagship: Claude Opus 4.7 (claude-opus-4-7). Current models: Opus 4.7 · Sonnet 4.6 · Haiku 4.5. Claude Opus 4.7 (claude-opus-4-7) is the current flagship as of April 16, 2026. Where this article references Opus 4.6 or earlier models, those references are historical. See current model tracker →. See current model tracker →

    Claude AI · Fitted Claude

    A Claude jailbreak is any technique designed to bypass Claude’s safety training and get it to produce content it would otherwise refuse. People search for this for different reasons — curiosity about how AI safety works, security research, or genuine attempts to exploit the model. Here’s what jailbreaking Claude actually looks like, why it’s harder than most people expect, and what happens when it does work.

    The honest framing: Claude is the most safety-hardened commercial AI model available in 2026. Standard jailbreak techniques have low single-digit success rates against it. That said, no model is unbreakable — persistent, multi-turn adversarial prompting has demonstrated real-world success. Anthropic publishes its research on this openly and updates defenses continuously.

    How Claude’s Safety System Works

    Claude’s safety isn’t a single content filter — it’s a layered defense built into the model at training time. Anthropic uses Constitutional AI, a technique where Claude is trained against a set of principles and learns to evaluate its own outputs. The model doesn’t just pattern-match on blocked keywords; it reasons about whether a response would cause harm given the full context of the request.

    On top of the trained model, Anthropic adds Constitutional Classifiers — a second layer that monitors inputs and outputs independently, trained on synthetic adversarial prompts across thousands of variations. Compared to an unguarded model, Constitutional Classifiers reduced the jailbreak success rate from 86% to 4.4% — blocking 95% of attacks that would otherwise bypass Claude’s built-in safety training.

    Common Jailbreak Techniques and Why They Don’t Work Well on Claude

    Persona injection (“DAN” / “do anything now”). Asking Claude to adopt an unrestricted persona — an “unfiltered AI,” a fictional character not bound by guidelines. Claude’s Constitutional AI training is robust against most direct persona injection attempts: the model declines the underlying request rather than complying through the fictional wrapper.

    Roleplay framing. Wrapping harmful requests in fictional or hypothetical scenarios — “write a story where a character explains how to…” Claude evaluates the real-world impact of its outputs, not just the fictional framing. A response that would cause harm outside fiction causes the same harm inside it.

    Token manipulation. Base64 encoding, unusual capitalization, Unicode substitution, and other character-level tricks to route requests past classifiers. Constitutional Classifiers are trained on these variations and handle most of them.

    Reasoning framing. Presenting harmful requests as academic, research, or security-related. Claude considers whether a request is plausibly legitimate given context — a genuine security research context differs from a claim of being a researcher with no supporting context.

    Where Jailbreaks Do Work

    The Mexico breach in early 2026 — where an attacker used over 1,000 Spanish-language prompts, role-playing Claude as an “elite hacker” in a fictional bug bounty program, eventually causing Claude to abandon its alignment context — demonstrated that persistent multi-turn escalation can work against even hardened models. The attack succeeded not through a clever single prompt but through sustained pressure, context manipulation, and gradual escalation across a long session.

    Multi-turn escalation still works at a non-trivial rate. Single-prompt jailbreaks are mostly defeated. Long sessions with gradual escalation remain a real vulnerability. Anthropic updated Claude Opus 4.6 with real-time misuse detection following the incident.

    Anthropic’s Public Red-Teaming Program

    Anthropic doesn’t just build defenses — it tests them publicly. Over 180 security researchers spent more than 3,000 hours over two months trying to jailbreak Claude using Constitutional Classifiers, offering a $15,000 bounty for a successful universal jailbreak. They weren’t able to find one during that period, though subsequent research has found partial techniques.

    This transparency is part of Anthropic’s approach: publish the research, run public bug bounties, and update defenses based on what adversaries discover. The Constitutional Classifiers paper is publicly available and describes the methodology in full.

    What Happens When Claude Gets Jailbroken

    The consequences range from producing harmful content (the worst case) to simply generating off-policy responses that violate Anthropic’s usage terms. Accounts used to jailbreak Claude are banned. In the Mexico case, Anthropic banned the implicated accounts and shipped defensive updates to the model within weeks of discovery.

    Using jailbreaks to extract harmful content violates Anthropic’s terms of service regardless of intent. Using jailbroken Claude to cause real-world harm — as in the Mexico case — is a criminal matter.

    The Practical Alternative to Jailbreaking

    Most people searching for jailbreaks actually want Claude to do something specific it’s currently refusing. Claude’s refusals are mostly a context problem, not a censorship problem. Providing more context about your role, purpose, and authorization frequently resolves apparent refusals that feel like hard limits. If you’re building a product that needs capabilities beyond what the consumer interface allows, the Claude API with appropriate operator system prompts is the legitimate path — not jailbreaking.

    For Claude’s full privacy and safety stance, see Is Claude Safe to Use? and Claude Privacy: What Anthropic Does With Your Data.

    Frequently Asked Questions

    Can Claude be jailbroken?

    Yes, but with difficulty. Standard single-prompt jailbreak techniques have very low success rates against Claude’s Constitutional AI training and Constitutional Classifiers. Persistent multi-turn escalation over long sessions has demonstrated real-world success. Anthropic continuously updates defenses and bans accounts used for jailbreaking.

    Is jailbreaking Claude illegal?

    Jailbreaking violates Anthropic’s terms of service. Using jailbreak techniques to cause real-world harm — breaching systems, generating CSAM, synthesizing weapons — is illegal regardless of the AI tool involved. Anthropic bans accounts and cooperates with law enforcement when illegal activity is discovered.

    Why does Claude refuse some requests that seem harmless?

    Claude evaluates requests as policies — imagining many different people making the same request and calibrating its response to the realistic distribution of intent. Some requests that are genuinely harmless get caught by this calibration. Providing more context about your specific purpose and role usually resolves these cases without needing to “jailbreak” anything.

    ⬡

    Deploying Claude for your organization?

    We configure Claude correctly — right plan tier, right data handling, right system prompts, real team onboarding. Done for you, not described for you.

    Learn about our implementation service →

    Need this set up for your team?
    Talk to Will →

  • Claude AI Privacy: What Anthropic Does With Your Conversations

    Claude AI Privacy: What Anthropic Does With Your Conversations

    Last refreshed: May 15, 2026

    Claude AI · Fitted Claude

    Before you paste anything sensitive into Claude, you should understand what Anthropic does with your conversations. The answer varies significantly by plan — and most people are on the plan with the least data protection. Here’s the complete picture.

    The key fact most people miss: On Free and Pro plans, Anthropic may use your conversations to train future Claude models. You can opt out in settings. Team and Enterprise plans have stronger protections and the Enterprise tier supports custom data handling agreements for regulated industries.

    Claude Data Handling by Plan

    Plan Training data use Human review possible? Custom data agreements
    Free Yes (opt-out available) Yes —
    Pro Yes (opt-out available) Yes —
    Team No (by default) Limited —
    Enterprise No Configurable ✓ BAA available

    How to Opt Out of Training Data Use

    On Free and Pro plans, you can disable conversation use for model training in your account settings. Go to Settings → Privacy → and toggle off “Help improve Claude.” This applies to future conversations — it doesn’t retroactively remove past conversations from training data already collected.

    What Anthropic Can See

    Anthropic employees may review conversations for safety research, model improvement, and trust and safety purposes. This applies to all plan tiers, though the scope and purpose of review is more restricted on Team and Enterprise. Human reviewers follow internal access controls, but if you’re sharing genuinely sensitive information, the better approach is to use Enterprise with appropriate data handling agreements — not to rely on the assumption that your specific conversation won’t be reviewed.

    Data Retention

    Anthropic retains conversation data for a period before deletion. The specific retention period isn’t published in a simple number — it varies based on account type and purpose. Your conversation history in the Claude.ai interface can be deleted by you at any time from Settings. Deletion from the UI doesn’t guarantee immediate removal from all backend systems, and may not remove data already used in training.

    Claude and GDPR

    For users in the EU, Anthropic operates under GDPR obligations. This includes rights to data access, correction, and deletion. Anthropic’s privacy policy covers these rights and how to exercise them. For organizations subject to GDPR with stricter requirements around AI data processing, Enterprise is the appropriate tier — it supports data processing agreements and more granular controls.

    What Not to Share With Claude on Standard Plans

    On Free or Pro plans, avoid sharing:

    • Patient health information (HIPAA-regulated)
    • Client confidential data under NDA
    • Non-public financial information
    • Personally identifiable information beyond what the task requires
    • Trade secrets or proprietary business processes

    For a full breakdown of Claude’s safety posture beyond just privacy, see Is Claude AI Safe? For current, authoritative terms, always refer to Anthropic’s privacy policy directly.

    Frequently Asked Questions

    Does Claude store your conversations?

    Yes. Anthropic retains conversation data for a period of time. You can delete your conversation history from the Claude.ai interface, but this doesn’t guarantee immediate removal from all backend systems or data already incorporated into training.

    Is Claude HIPAA compliant?

    Not on standard plans. HIPAA compliance requires a Business Associate Agreement (BAA) with Anthropic, which is only available on the Enterprise plan. Do not share patient health information with Claude on Free, Pro, or Team plans.

    Can I stop Anthropic from using my conversations to train Claude?

    Yes, on Free and Pro plans you can opt out in Settings → Privacy. Team plans don’t use conversations for training by default. On Enterprise, this is governed by your data processing agreement.

    Is Claude private?

    Claude conversations are not end-to-end encrypted in the way messaging apps are. Anthropic can access conversation data. “Private” in the sense of not being shared with third parties — yes, Anthropic doesn’t sell your data. Private in the sense of completely inaccessible to the company that runs it — no.

    ⬡

    Deploying Claude for your organization?

    We configure Claude correctly — right plan tier, right data handling, right system prompts, real team onboarding. Done for you, not described for you.

    Learn about our implementation service →

    Need this set up for your team?
    Talk to Will →

  • Is Claude AI Safe? Data Handling, Content Safety, and What to Know

    Is Claude AI Safe? Data Handling, Content Safety, and What to Know

    Last refreshed: June 9, 2026

    Claude AI · Fitted Claude

    Claude is built by Anthropic — a company whose stated mission is AI safety. But “safe” means different things depending on what you’re asking: Is Claude safe to use with sensitive information? Is it safe for children? Does it produce harmful content? Is it psychologically safe to rely on? Here’s the honest answer to each version of the question.

    Short answer: Claude is one of the safest AI assistants available for general professional use. It’s designed to refuse harmful requests, be honest about uncertainty, and avoid manipulation. For sensitive business data, read the data handling section below before sharing anything confidential.

    Is Claude Safe to Use? By Use Case

    Concern Safety Level Notes
    General professional use ✅ Safe Standard writing, research, analysis
    Children and minors ⚠️ Use with awareness Claude declines adult content but isn’t a parental control tool
    Sensitive personal information ⚠️ Read privacy policy Conversations may be used to improve models on free/Pro tiers
    Confidential business data ⚠️ Enterprise tier recommended Enterprise has stronger data handling commitments
    HIPAA-regulated data ❌ Not on standard plans Requires Enterprise with a BAA from Anthropic
    Harmful content generation ✅ Declines Claude refuses instructions for weapons, self-harm, etc.

    How Anthropic Builds Safety Into Claude

    Anthropic uses a training methodology called Constitutional AI — Claude is trained against a set of principles rather than purely optimizing for user approval. This means Claude is more likely to push back on bad premises, decline harmful requests, and express uncertainty rather than generate a confident-sounding wrong answer.

    Concretely: Claude won’t provide instructions for creating weapons, won’t generate content that sexualizes minors, won’t help with clearly illegal activities targeting individuals, and is designed to be honest rather than sycophantic. These are trained behaviors, not just content filters bolted on afterward.

    Data Safety: What Happens to Your Conversations

    This is the area that matters most for professional users. Anthropic’s data handling varies by plan:

    Free and Pro plans: Conversations may be used by Anthropic to improve Claude’s models. You can opt out of this in your account settings. Anthropic retains conversation data for a period before deletion.

    Team plan: Stronger data handling commitments. Conversations are not used to train models by default.

    Enterprise plan: Custom data handling agreements available. This is the tier for organizations with compliance requirements — HIPAA, SOC 2, GDPR, etc. A Business Associate Agreement (BAA) from Anthropic is required before sharing any HIPAA-regulated data.

    For current, authoritative data handling details, check Anthropic’s privacy policy directly — it supersedes any summary here. For privacy-specific questions, see Claude AI Privacy: What Anthropic Does With Your Data.

    Is Claude Psychologically Safe?

    Claude is designed not to manipulate users, not to foster unhealthy dependency, and not to tell people what they want to hear at the expense of accuracy. It will disagree with you, push back on flawed premises, and decline to validate bad decisions. Whether that’s “safe” depends on your frame — but it’s a deliberate design choice that makes Claude more honest and less likely to be weaponized as a validation machine.

    Frequently Asked Questions

    Is Claude AI safe to use?

    Yes, for general professional use. Claude is designed to refuse harmful requests, be honest, and avoid manipulation. For sensitive business data or regulated information, review Anthropic’s data handling policies for your plan tier before sharing anything confidential.

    Is Claude safe for children?

    Claude declines to generate adult or harmful content, which makes it safer than many AI tools. However, it’s not a purpose-built parental control system and shouldn’t be treated as one. Anthropic’s Terms of Service require users to be 18 or older, or to have parental permission.

    Can I share confidential business information with Claude?

    On standard plans (Free, Pro), conversations may be reviewed by Anthropic and used for model improvement. For confidential business data, use the Team or Enterprise plan — Enterprise offers custom data handling agreements. Never share HIPAA-regulated data without a Business Associate Agreement in place.

    Is Claude safer than ChatGPT?

    Both Claude and ChatGPT have safety measures in place. Claude’s Constitutional AI training approach is designed specifically around safety as a core methodology rather than an add-on. For data handling, the comparison depends on which plan tier you’re on for each product — Enterprise tiers of both have stronger commitments than free or standard paid plans.

    ⬡

    Deploying Claude for your organization?

    We configure Claude correctly — right plan tier, right data handling, right system prompts, real team onboarding. Done for you, not described for you.

    Learn about our implementation service →

    Is Claude safe to use for sensitive or confidential work?

    For highly sensitive work, use the Claude API (data not stored by default) or Claude Enterprise (contractual data protections, no training on your data). The standard claude.ai consumer plans store conversations and may use them for model improvement unless you opt out. Never send passwords, API keys, or financial account numbers to any AI system.

    Does Claude have content filters and safety guardrails?

    Yes. Claude is trained with Constitutional AI and RLHF to decline harmful requests, avoid generating dangerous content, and flag requests that violate Anthropic’s usage policies. Claude’s safety posture is conservative by default.

    Can Claude be used safely with children?

    Claude has safety guardrails that prevent it from producing inappropriate content for minors. Educational platforms deploying Claude for K-12 use must comply with COPPA, FERPA, and Anthropic’s usage policies, and should use Enterprise agreements with appropriate data protections.

    Need this set up for your team?
    Talk to Will →

  • Dario Amodei: CEO of Anthropic and the Future of AI Safety

    Dario Amodei: CEO of Anthropic and the Future of AI Safety

    Last refreshed: May 15, 2026

    Claude AI · Fitted Claude

    Dario Amodei is the CEO and co-founder of Anthropic, the AI safety company behind Claude. His trajectory — Princeton physics, Stanford PhD, OpenAI VP of Research, then Anthropic founder — traces the arc of modern AI development. Forbes estimated his net worth at $7 billion as of February 2026, reflecting his co-founder equity as Anthropic approaches a potential IPO.

    Early Life and Education

    Dario Amodei grew up in a family with deep intellectual roots — his father is a physician, his mother a chemist. He studied physics at Princeton University before earning a PhD in computational neuroscience at Stanford, where he researched the intersection of neural computation and machine learning. The neuroscience background proved directly relevant: understanding how biological neural networks process information informed his later work on understanding artificial ones.

    Career at OpenAI

    Amodei joined OpenAI in 2016 as a research scientist and rose to become Vice President of Research — one of the most senior technical roles in the organization during the period when OpenAI produced GPT-2, GPT-3, and early versions of DALL-E. His tenure coincided with OpenAI’s most productive research period and its transition from a pure research organization to a company with significant commercial ambitions.

    By 2021, Amodei and a group of colleagues had grown increasingly concerned that OpenAI’s commercial trajectory — particularly its deepening partnership with Microsoft — was creating tensions with rigorous AI safety research. The concerns were not primarily about OpenAI’s intentions but about whether a company under those commercial pressures could systematically prioritize safety as its primary obligation.

    Co-Founding Anthropic

    In 2021, Amodei led the founding of Anthropic alongside his sister Daniela Amodei, Jared Kaplan, Chris Olah, Tom Brown, Sam McCandlish, and Jack Clark. The company was structured as a public benefit corporation — a legal form that formally embeds the safety mission into its governing documents, creating accountability beyond a standard corporate charter.

    Amodei has consistently articulated a position that sits between AI pessimism and uncritical optimism: he believes advanced AI poses genuine existential-level risks, and that the way to address those risks is not to slow development but to pursue it more carefully, with safety research as the primary scientific agenda rather than an afterthought.

    Leadership Style and Public Profile

    Amodei is more publicly visible than most AI lab CEOs, regularly writing long-form essays on AI policy and safety, appearing before Congress, and engaging directly with critics of both the AI safety field and of Anthropic specifically. His October 2024 essay “Machines of Loving Grace” — a detailed argument for why advanced AI could be profoundly beneficial — generated significant attention and debate across the AI community.

    Net Worth

    Forbes estimated Dario Amodei’s net worth at approximately $7 billion as of February 2026, reflecting his co-founder equity in Anthropic at the company’s current valuation. As one of the largest individual stakeholders in a company targeting a $400-500B IPO valuation, this figure could change substantially if the public offering proceeds as expected.

    Frequently Asked Questions

    What is Dario Amodei’s net worth?

    Forbes estimated approximately $7 billion as of February 2026, based on his co-founder equity in Anthropic.

    Why did Dario Amodei leave OpenAI?

    Amodei and colleagues grew concerned that commercial pressures — particularly OpenAI’s Microsoft partnership — were creating structural tensions with rigorous AI safety research as the primary mission.

    Where did Dario Amodei go to school?

    Dario Amodei studied physics at Princeton and earned a PhD in computational neuroscience from Stanford University.

  • Anthropic IPO 2026: What’s Confirmed, What’s Rumored, and Where to Track It

    Anthropic IPO 2026: What’s Confirmed, What’s Rumored, and Where to Track It

    Last refreshed: May 15, 2026

    ⚠️ No confirmed IPO date exists as of May 8, 2026. Anthropic has not filed an S-1, set a ticker, or announced a listing date. What exists are credible reports of a Q4 2026 target — but no official confirmation. Everything below is sourced and dated. Click any link to get the latest.

    Where Things Actually Stand

    Anthropic is widely expected to pursue an IPO, and the signals are real — but no date has been set. Here is what is confirmed versus what is reported:

    Confirmed Facts (Primary Sources)

    • Current valuation: $380 billion — set in the February 2026 Series G round led by GIC and Coatue. This is the last confirmed, announced valuation. (CNBC, April 29 2026)
    • Revenue run rate: $30B+ annualized — confirmed by Anthropic directly in May 2026. Sources with knowledge of financials put the real figure closer to $40B. (TechCrunch, April 29 2026)
    • IPO law firm engaged: Wilson Sonsini hired to prepare for a potential public listing — confirmed by the Financial Times in December 2025.
    • Preliminary bank conversations: Anthropic has held early-stage talks with investment banks — confirmed by multiple sources, no banks named publicly.
    • No S-1 filed. The SEC has received no public filing from Anthropic as of this writing.

    Reported But Unconfirmed

    • Q4 2026 IPO target — discussed by Anthropic executives internally according to The Information. Bankers reportedly expect the offering could raise more than $60 billion. (TECHi, sourcing The Information)
    • ~$900 billion valuation round in progress — as of April 30, 2026, TechCrunch reported Anthropic was asking investors to submit allocations within 48 hours for a ~$50 billion raise at a $850–$900 billion valuation. A board decision was expected in May 2026. Anthropic declined to comment. (TechCrunch, April 30 2026)
    • October 2026 — cited in some reports as the earliest possible listing window. Not confirmed by Anthropic.
    • $60B+ raise — reported figure for the eventual IPO offering size. Unconfirmed.

    The Valuation Trajectory

    The speed of Anthropic’s private-market repricing is unlike anything in recent tech history:

    • March 2025: $61.5 billion (Series D, led by Lightspeed)
    • September 2025: $183 billion (Series F)
    • February 2026: $380 billion (Series G, led by GIC and Coatue)
    • May 2026: ~$900 billion reportedly under discussion — not yet closed

    Some early backers are reportedly skipping the current round specifically to wait for IPO pricing — a signal that sophisticated money sees the public listing as potentially more attractive than another late-stage private markup.

    Why There’s No Confirmed Date Yet

    Anthropic is a public benefit corporation, which adds governance complexity to any listing. The company is also in the middle of closing what may be its final private round — and closing a $50 billion raise takes time. Until an S-1 is filed with the SEC, no IPO date is official. PitchBook analyst Kyle Stanford has noted that a crowded private financing cycle could push a listing into 2027 if the current round takes longer than expected.

    Who Owns Anthropic Before Any IPO

    Major confirmed investors include Amazon (up to $50 billion committed), Google (up to $40 billion committed), Nvidia ($30 billion), SoftBank ($30 billion), plus Accel, BlackRock-affiliated funds, Fidelity, General Catalyst, Goldman Sachs Alternatives, JPMorganChase, Lightspeed, Menlo Ventures, Morgan Stanley Investment Management, Sequoia, and Temasek. More than 1,000 enterprise customers now spend over $1 million annually on Claude — a figure Anthropic disclosed publicly in May 2026.

    Keep Up With This Story

    This is a fast-moving situation. The sources below are updated in real time — bookmark them if you want the latest as it breaks:

    Want the deeper picture on who is building this company? Read our analysis of Anthropic’s founders and leadership — the most-read piece on this site in this category.

  • Daniela Amodei: Co-Founder and President of Anthropic

    Daniela Amodei: Co-Founder and President of Anthropic

    Daniela Amodei is the President and co-founder of Anthropic, the AI safety company behind Claude. While her brother Dario Amodei serves as CEO and is the more publicly visible figure, Daniela runs the operational, commercial, and go-to-market sides of one of the most consequential AI companies in the world. She is, in practical terms, the reason Anthropic functions as a business.

    Quick facts: Daniela Amodei — President and co-founder of Anthropic. Previously VP of Operations at OpenAI. Before that: Stripe, Ropes & Gray. Co-founded Anthropic in 2021 with her brother Dario and five other former OpenAI researchers. Responsible for Anthropic’s business operations, sales, partnerships, and go-to-market strategy.

    Who Is Daniela Amodei?

    Daniela Amodei is the President of Anthropic, the AI safety company she co-founded in 2021 alongside her brother Dario Amodei and a group of senior researchers who departed OpenAI together. While Dario leads research and product as CEO, Daniela leads everything that keeps the company running as a viable business: revenue, partnerships, hiring, operations, and the commercial strategy behind Claude.

    She is among the most powerful operators in the AI industry — not a figurehead co-founder, but the executive who built Anthropic’s commercial foundation from zero while the research team focused on the models.

    Background and Career Before Anthropic

    Before Anthropic, Daniela spent years in operational and business roles that would prove directly relevant to building a fast-moving AI company from scratch.

    She attended Dartmouth College, where she studied economics. Her early career included a position at Ropes & Gray, a prominent law firm, before moving into the technology sector. She joined Stripe — the payments infrastructure company — where she worked in business operations during a period of significant growth for the company.

    The pivotal move came when she joined OpenAI as VP of Operations. She was one of the senior leaders who left OpenAI in 2020 and 2021 along with her brother Dario to found Anthropic. That cohort included several of OpenAI’s most senior researchers and operators, making it one of the most significant team departures in AI industry history.

    Role at Anthropic

    As President, Daniela’s domain at Anthropic covers the business side of the company end to end. Where Dario focuses on research direction, safety philosophy, and model development, Daniela owns:

    • Revenue and commercial growth — enterprise sales, partnerships, and the Claude business
    • Go-to-market strategy — how Anthropic positions and sells Claude to individuals, developers, and enterprises
    • Operations — the internal systems and processes that let a growing AI company function
    • Partnerships — major deals including Anthropic’s relationship with Amazon Web Services, one of the largest infrastructure commitments in AI company history
    • Hiring and team building — scaling the organization while maintaining culture

    The division of labor between Daniela and Dario mirrors a pattern common in successful tech companies: one founder focused on product and technology, one focused on the business that makes the technology sustainable. At Anthropic, that structure is unusually clean and appears to function well.

    Daniela Amodei and the Amazon Partnership

    One of the most significant commercial milestones under Daniela’s leadership as President was securing Anthropic’s partnership with Amazon Web Services. Amazon committed to invest up to $4 billion in Anthropic, with Claude models made available through AWS’s Bedrock platform. This deal established Anthropic’s commercial credibility and gave it the infrastructure scale to compete with OpenAI and Google DeepMind.

    Partnerships of this scale require sustained executive relationships and months of commercial negotiation — the kind of work that falls squarely in Daniela’s domain.

    The Amodei Siblings Running Anthropic

    The dynamic between Daniela and Dario Amodei at Anthropic is worth understanding because it’s unusual. Co-founders who are siblings and who have distinct, non-overlapping domains are relatively rare. In most tech companies, co-founders compete for influence. At Anthropic, the operational split appears deliberate and functional: Dario owns the mission and the models, Daniela owns the machine that funds the mission.

    Dario has spoken publicly about AI safety, the risks of powerful AI systems, and Anthropic’s research philosophy. Daniela tends to operate more quietly — she is less frequently the face of Anthropic in press interviews but is consistently present in the company’s major commercial announcements and partnership moments.

    Net Worth and Anthropic’s Valuation

    Anthropic has raised billions of dollars in venture funding from investors including Google, Amazon, and Spark Capital, with valuations that have grown significantly through each funding round. As a co-founder and President holding equity in the company, Daniela Amodei’s net worth is tied primarily to Anthropic’s private valuation.

    Anthropic is not publicly traded, so precise figures are not available. At the company’s reported valuations, co-founders with meaningful equity stakes hold substantial paper wealth — though the actual liquidity of that wealth depends on if and when Anthropic conducts an IPO or secondary transactions.

    Why Daniela Amodei Matters for Claude

    Claude exists because Anthropic exists as a viable company. Daniela Amodei is one of the primary reasons Anthropic is viable. The research team can build frontier AI models, but without a functioning commercial operation those models don’t reach users, don’t generate revenue, and don’t fund the next generation of research.

    Every enterprise Claude deployment, every API integration, every AWS customer using Claude through Bedrock, every API integration, every AWS customer using Claude through Bedrock — these exist in part because of the commercial infrastructure Daniela has built. The Claude you use is as much a product of her work as it is of the research team’s.

    Frequently Asked Questions

    Who is Daniela Amodei?

    Daniela Amodei is the President and co-founder of Anthropic, the AI company behind Claude. She previously served as VP of Operations at OpenAI before co-founding Anthropic in 2021 with her brother Dario Amodei and other former OpenAI researchers.

    Is Daniela Amodei related to Dario Amodei?

    Yes. Daniela and Dario Amodei are siblings. Dario is the CEO of Anthropic; Daniela is the President. They co-founded Anthropic together in 2021 along with five other former OpenAI researchers.

    What does Daniela Amodei do at Anthropic?

    As President, Daniela oversees Anthropic’s business operations, commercial strategy, revenue, partnerships, and go-to-market. She is responsible for the business side of Anthropic while Dario leads research and product.

    Where did Daniela Amodei work before Anthropic?

    Before co-founding Anthropic, Daniela was VP of Operations at OpenAI. Prior to OpenAI she worked at Stripe in business operations, and earlier in her career she was at the law firm Ropes & Gray. She studied economics at Dartmouth College.

    What is Daniela Amodei’s net worth?

    Daniela Amodei’s net worth is not publicly known — Anthropic is a private company and does not disclose individual equity stakes. Her net worth is tied primarily to her equity in Anthropic, which has been valued at billions of dollars across successive funding rounds from investors including Amazon and Google.




  • The Real Monthly Cost of Running Claude Managed Agents 24/7

    The Real Monthly Cost of Running Claude Managed Agents 24/7

    Tygart Media Strategy
    Volume Ⅰ · Issue 04Quarterly Position
    By Will Tygart
    • Long-form Position
    • Practitioner-grade

    If you’re considering running Claude Managed Agents around the clock, you want a number. Not “it depends.” An actual number you can put in a budget. Here’s the math, worked out by scenario, with the honest caveats about where the real costs are.

    The Formula

    Total monthly cost = (Active session hours × $0.08) + token costs + optional tool costs

    The $0.08/session-hour charge only applies during active execution. Idle time — waiting for input, tool confirmations, external API responses — doesn’t count. This matters significantly for 24/7 workloads, because very few agents are active 100% of the time even when “running around the clock.”

    The Maximum Theoretical Cost

    Scenario: Agent running continuously, zero idle time, 24 hours a day, 30 days a month.

    • Session runtime: 24 hrs × $0.08 × 30 days = $57.60/month
    • Token costs: separate, highly variable (see below)

    $57.60/month is the ceiling on session runtime charges. You cannot pay more than this in session fees under any 24/7 scenario. But here’s the reality: that ceiling assumes zero idle time across the entire month, which doesn’t describe any real production agent.

    Realistic 24/7 Scenarios

    Monitoring Agent (High Idle Ratio)

    Runs continuously watching for triggers — error alerts, specific data patterns, incoming requests. Activates on trigger, processes, returns to monitoring state.

    • Assumption: 5% active execution time (watching 95% of the time, executing 5%)
    • Active hours: 24 × 30 × 0.05 = 36 hours/month
    • Session runtime: 36 × $0.08 = $2.88/month
    • Token costs: low — moderate bursts on trigger events
    • Realistic total: $5–15/month

    Customer Support Agent (Business Hours Active)

    “24/7” in the sense of always-available, but actual request volume concentrates in business hours. Waits for tickets, processes them, waits again.

    • Assumption: 8 hours/day active execution, 16 hours waiting
    • Active hours: 8 × 30 = 240 hours/month
    • Session runtime: 240 × $0.08 = $19.20/month
    • Token costs: depends heavily on ticket volume and average length
    • At 100 tickets/day with moderate length: likely $30–80/month in tokens
    • Realistic total: $50–100/month

    Continuous Autonomous Pipeline

    Batch processing agent that runs continuously through a queue with minimal waiting — the closest to true 24/7 active execution.

    • Assumption: 20 hours/day truly active (4 hours queue exhaustion/maintenance)
    • Active hours: 20 × 30 = 600 hours/month
    • Session runtime: 600 × $0.08 = $48/month
    • Token costs: high — continuous processing means continuous token consumption
    • This is where tokens become the dominant cost driver by a significant margin
    • Realistic total: $200–500+/month (tokens dominate)

    The Real Variable: Token Costs

    For any 24/7 workload that’s genuinely busy, token costs will substantially exceed session runtime costs. The math:

    A moderately active agent processing 10,000 input tokens and 2,000 output tokens per hour with Claude Sonnet 4.6:

    • Input: 10,000 tokens × $3/million = $0.03/hour
    • Output: 2,000 tokens × $15/million = $0.03/hour
    • Token cost: $0.06/hour vs. session runtime of $0.08/hour — roughly equal at this volume

    Scale to 100,000 input tokens and 20,000 output tokens per hour (a busy processing agent):

    • Input: $0.30/hour; Output: $0.30/hour
    • Token cost: $0.60/hour vs. session runtime of $0.08/hour — tokens are 7.5× the runtime charge

    The session runtime fee is flat and bounded. Token costs scale with workload volume. For high-volume 24/7 agents, optimize token efficiency (prompt caching, context management, output brevity) before worrying about the session runtime charge.

    Prompt Caching Changes the Token Math

    If your agent has a large, stable system prompt — common in agents with extensive tool definitions or knowledge bases — prompt caching dramatically reduces input token costs. Cache hits cost a fraction of base input rates. For a 24/7 agent with a 20,000-token system prompt hitting the same context repeatedly, caching that prompt can cut input costs by 80–90%. The session runtime charge is unchanged, but the total cost picture improves significantly.

    The Budget Summary

    Agent Type Runtime/mo Typical Total
    Monitoring / low activity ~$3 $5–15
    Support agent (business hours volume) ~$19 $50–100
    Continuous processing pipeline ~$48 $200–500+
    Theoretical maximum (zero idle) $57.60 Unbounded (tokens)

    Complete pricing reference: Claude Managed Agents Pricing Guide. How idle time affects billing: Idle Time and Billing Explained. All questions: FAQ Hub.

    What to do next

    Now that you have the cost math — here’s how to choose and implement

    You now know what Managed Agents costs at scale. The next decision is whether it’s the right architecture vs. OpenAI’s equivalent — and what the implementation actually looks like in practice.

  • Claude Managed Agents vs. OpenAI Agents API — A Direct Comparison

    Claude Managed Agents vs. OpenAI Agents API — A Direct Comparison

    TL;DR — Pick one in 30 seconds

    Choose Claude Managed Agents for zero-infra, fast production deployment. Choose OpenAI Agents API if you need multi-model flexibility or already run on OpenAI infrastructure.

    Feature Claude Managed Agents OpenAI Agents API
    Model lock-in Claude only GPT-4o, o3 — OAI only
    Setup complexity Zero infra — fully managed SDK — you build the harness
    Memory Built-in (public beta, May 2026) Manual via vector DB
    Multiagent Native (lead + specialists) Swarm/SDK patterns
    Pricing $0.08/session-hr + tokens Token-only (no session fee)
    Best for Fast production, Claude-native Multi-model, existing OAI infra

    Model Accuracy Note — Updated May 2026

    Current flagship: Claude Opus 4.7 (claude-opus-4-7). Current models: Opus 4.7 · Sonnet 4.6 · Haiku 4.5. Claude Opus 4.6 referenced in this article has been superseded. See current model tracker →

    Tygart Media Strategy
    Volume Ⅰ · Issue 04Quarterly Position
    By Will Tygart
    • Long-form Position
    • Practitioner-grade

    You’re evaluating hosted agent infrastructure. Both Anthropic and OpenAI have one. Before you commit to either, here’s what’s actually different — not the marketing version, the architectural and pricing version.

    Bottom Line Up Front

    If your stack is Claude-native and you want to get to production fast without building orchestration infrastructure, Managed Agents is hard to beat. If you need multi-model flexibility or have OpenAI deeply embedded in your stack, the calculus changes. Lock-in is real on both sides.

    Still Deciding?

    I’ve run both. Email me your use case and I’ll tell you which one fits.

    No pitch. If Claude isn’t the right call for what you’re building, I’ll tell you that too.

    Email Will → will@tygartmedia.com

    What Each Product Is

    Claude Managed Agents

    Anthropic’s hosted runtime for long-running Claude agent work. You define an agent (model, system prompt, tools, guardrails), configure a cloud environment, and launch sessions. Anthropic handles sandboxing, state management, checkpointing, tool orchestration, and error recovery. Launched April 8, 2026 in public beta.

    OpenAI Agents API

    OpenAI’s hosted agent infrastructure layer, launched earlier in 2026. Provides similar capabilities: hosted execution, tool integration, multi-agent coordination. Supports multiple OpenAI models (GPT-4o, o1, o3, etc.).

    Model Flexibility

    Managed Agents: Claude models only. Sonnet 4.6 and Opus 4.6 are the primary options for agent work. No multi-model mixing within the managed infrastructure.

    OpenAI Agents API: OpenAI models only, but a wider current model lineup (GPT-4o, o1, o3-mini depending on task). Also Claude-only within its own ecosystem — not multi-model in the cross-provider sense.

    The practical implication: If your evaluation is “I want the best model for this specific task regardless of provider,” neither hosted solution gives you that. Both lock you to their provider’s models. The multi-model comparison matters for self-hosted frameworks (LangChain, etc.), not for managed hosted solutions.

    Pricing Structure

    Claude Managed Agents: Standard Claude token rates + $0.08/session-hour of active runtime. Idle time doesn’t bill. Code execution containers included in session runtime — not separately billed.

    OpenAI Agents API: Standard OpenAI token rates + usage-based tooling costs. Pricing structure varies by tool and model tier. Verify current rates at OpenAI’s pricing page — rates have changed multiple times as their agent products have evolved.

    Direct comparison difficulty: Without modeling the same specific workload against both providers’ current rates, headline comparisons mislead. Token rates differ by model, model capabilities differ, and “session runtime” isn’t a category OpenAI uses. Model the workload, not the headline number.

    Infrastructure and Lock-In

    Both solutions create meaningful lock-in. This isn’t a criticism — it’s an honest description of the trade-off you’re making:

    Claude Managed Agents lock-in: Your agents run on Anthropic’s infrastructure with their tools, session format, sandboxing model, and checkpointing. Migrating to OpenAI’s Agents API or self-hosted infrastructure requires rearchitecting session management, tool integrations, and guardrail logic. One developer’s reaction at launch: “Once your agents run on their infra, switching cost goes through the roof.”

    OpenAI Agents API lock-in: Symmetric. Same dynamic in reverse. OpenAI’s session format, tool integration patterns, and infrastructure assumptions create equivalent switching costs to move to Anthropic’s platform.

    The honest framing: You’re not choosing “open” vs. “locked.” You’re choosing which provider’s lock-in you’re more comfortable with, given your existing infrastructure, model preferences, and vendor relationship.

    Data Sovereignty

    Both solutions run your data on provider-managed infrastructure. Neither currently offers native on-premise or multi-cloud deployment for the managed hosted layer. For companies with strict data sovereignty requirements, this is a parallel constraint on both platforms — not a differentiator.

    Production Track Record

    Claude Managed Agents: Launched April 8, 2026. Production users at launch: Notion, Asana, Rakuten (5 agents in one week), Sentry, Vibecode, Allianz. Anthropic’s agent developer segment run-rate exceeds $2.5 billion.

    OpenAI Agents API: Earlier launch gives more time in production, but the product has been revised significantly since initial release. Longer production history, but also more legacy architectural assumptions baked in.

    When to Choose Claude Managed Agents

    • Your stack is already Claude-native (you’re using Sonnet or Opus for most model calls)
    • You want to reach production without building orchestration infrastructure
    • Your tasks are long-running and asynchronous — the session-hour model fits naturally
    • The Notion, Asana, or Sentry integrations are relevant to your workflow
    • You want Anthropic’s specific safety and reliability guarantees

    When to Consider OpenAI’s Agents API Instead

    • Your stack is already heavily OpenAI-integrated (GPT-4o for primary model work, existing tool integrations)
    • You need access to reasoning models (o1, o3) for specific task types — Anthropic’s equivalent is Claude’s extended thinking, which has different characteristics
    • The specific tool integrations in OpenAI’s ecosystem are better matched to your stack
    • You want more production time at scale before committing to a platform

    When to Use Neither (Self-Hosted Frameworks)

    LangChain, LlamaIndex, and similar self-hosted frameworks remain viable — and better — when you genuinely need multi-model flexibility, on-premise execution, or tighter loop control than either hosted solution provides. The trade-off is engineering effort: months of infrastructure work that Managed Agents or OpenAI’s API eliminates.

    Complete pricing breakdown: Claude Managed Agents Pricing Reference. All Managed Agents questions: FAQ Hub. Enterprise deployment example: Rakuten: 5 Agents in One Week.