Meta tags:
description=Large language models are very good at giving people what they ask for, which can also lead to some problems. They are people-pleasers with excellent syntax, and they can produce code that compiles, runs, and looks right in a pull request. That surface-level validity creates confidence before the system has earned it. The risk is not that AI writes broken code. The risk is;

Headings (most frequently used words):

the, webinars, fast, and, devopsdigest, main, menu, related, links, industry, news, email, signup, search, form, upcoming, on, demand, analyst, reports, white, papers, media, partners, user, login, latest, hot, topics, exploitable, ownership, problem, automation, has, limits, what, productivity, means, now, where, vibe, coding, fits, move, ask, better, questions,

Text of the page (most frequently used words):
the (86), and (48), 2026 (46), code (38), for (30), that (23), june (22), software (22), not (18), security (17), july (16), with (13), testing (12), has (11), 2025 (11), how (10), more (10), whether (10), development (9), generated (9), are (9), announced (9), platform (8), systems (8), risk (8), teams (8), real (8), but (8), does (8), coding (7), api (7), gartner (7), enterprise (7), new (7), what (7), agentic (7), data (7), test (7), can (7), tools (7), source (6), cloud (6), isn (6), your (6), faster (6), report (6), magic (6), quadrant (6), production (6), system (6), open (5), devops (5), agent (5), hybrid (5), speed (5), still (5), about (5), cyber (5), gitlab (5), they (5), ibm (5), when (5), work (4), engineering (4), devsecops (4), integration (4), why (4), developer (4), only (4), delivery (4), productivity (4), cost (4), validate (4), devopsdigest (4), all (4), critical (4), point (4), mesh (4), gigaom (4), radar (4), platforms (4), network (4), design (4), webinars (4), cdata (4), launched (4), developers (4), releases (4), mcp (4), agents (4), automated (4), rapidfort (4), dependency (4), southworks (4), ownership (4), review (4), because (4), fails (4), vibe (3), quality (3), monitoring (3), digital (3), analytics (3), need (3), right (3), untested (3), means (3), every (3), gets (3), built (3), ready (3), building (3), organizations (3), service (3), helps (3), velocity (3), build (3), blog (3), forum (3), papers (3), guide (3), best (3), practices (3), googletest (3), safety (3), application (3), scale (3), insights (3), from (3), events (3), check (3), reports (3), firewall (3), email (3), industry (3), compliance (3), generate (3), news (3), connect (3), free (3), autonomous (3), baz (3), between (3), automatically (3), entire (3), snaplogic (3), general (3), availability (3), shipping (3), cequence (3), changes (3), perforce (3), around (3), unisys (3), partner (3), reversinglabs (3), ask (3), hard (3), assumptions (3), own (3), stay (3), looks (3), under (3), fast (3), business (3), however (3), where (3), become (3), model (3), problem (3), cheaper (3), transformation (2), dev (2), native (2), automation (2), powered (2), end (2), zero (2), models (2), sdlc (2), now (2), rewriting (2), level (2), most (2), enterprises (2), one (2), age (2), engineers (2), beyond (2), gateways (2), next (2), virtualization (2), gap (2), request (2), password (2), sponsor (2), guidelines (2), vendor (2), global (2), white (2), waf (2), comparison (2), world (2), innovation (2), based (2), performance (2), web (2), 000 (2), range (2), analyst (2), surface (2), miercom (2), trust (2), making (2), transcend (2), exploring (2), starting (2), llm (2), analysis (2), pull (2), fixes (2), sase (2), access (2), policy (2), without (2), downstream (2), services (2), applications (2), edition (2), python (2), sdk (2), cli (2), launches (2), azul (2), jvm (2), vulnerability (2), assessment (2), increasingly (2), sembi (2), testrail (2), xray (2), leader (2), planner (2), gateway (2), through (2), eliminate (2), before (2), released (2), builder (2), into (2), harness (2), worker (2), writing (2), users (2), major (2), updates (2), intelligence (2), assisted (2), natural (2), language (2), unified (2), updated (2), couchbase (2), plane (2), its (2), distributed (2), git (2), instead (2), strategic (2), partnership (2), antenna (2), independent (2), third (2), party (2), deliver (2), deloitte (2), red (2), hat (2), collaboration (2), libraries (2), proven (2), process (2), validation (2), bob (2), multi (2), capabilities (2), use (2), johnny (2), halife (2), part (2), cannot (2), away (2), change (2), someone (2), questions (2), outcome (2), architecture (2), deliberate (2), needs (2), better (2), increases (2), deployment (2), just (2), across (2), product (2), generation (2), useful (2), experimentation (2), interactions (2), should (2), those (2), human (2), limits (2), ship (2), failure (2), technical (2), debt (2), time (2), implementations (2), tell (2), you (2), passes (2), will (2), checks (2), guardrails (2), bad (2), also (2), creates (2), same (2), obvious (2), during (2), writes (2), good (2), very (2), people (2), produce (2), main (2), vsm, tokenmaxxing, serverless, remote, mobile, mlops, mainframe, low, iac, commerce, devex, culture, database, dataops, containers, k8s, bizdevops, apm, aiops, agile, hot, topics, reactive, driven, observability, defect, itops, sized, bigger, entering, phase, expansion, competitive, realignment, working, feeling, talent, crunch, great, experience, accelerating, attacked, governing, abundance, tighter, visa, scrutiny, extraordinary, ability, case, miss, risks, outpaced, measure, autoscaling, smart, scheduling, frontier, customer, person, their, hidden, misconfigurations, crisis, undermining, closing, genai, tooling, while, rely, manual, latest, username, user, login, contact, editor, tips, rules, editorial, program, sponsors, navigating, rss, feed, media, partners, state, project, economics, essential, environment, maximize, functional, lean, strategy, adoption, challenges, intelligent, era, securing, illumio, attack, management, forrester, wave, anti, phishing, assistants, frost, sullivan, recognition, benchmark, certified, true, potential, appsec, actionable, achieve, cra, secure, realistic, dependencies, client, escaping, regression, slowdowns, long, lived, impact, explained, mocks, autonomously, pipeline, improving, embedded, clion, parasoft, smarter, requests, tests, validates, java, identity, internet, optimization, tuning, virtual, event, bridging, unit, microservices, waiting, demand, august, continuous, delivering, amazing, experiences, upcoming, search, form, sign, signup, three, products, address, blind, spot, exploitation, able, find, announces, prioritization, been, named, recognized, fourth, consecutive, year, introduced, sits, codebase, routing, idea, dynamic, loops, instantly, detect, root, cause, patch, vulnerabilities, bugs, proactively, plans, classes, issues, reach, template, capability, operationalize, turning, existing, pipelines, safely, run, handle, release, fundamentally, interact, agnostic, execution, infrastructure, layer, preview, letting, host, repositories, single, central, provider, embed, benchmarks, directly, solutions, help, protect, supply, chain, against, threats, announce, lightwell, package, catalog, combine, curation, hardening, spectra, assure, including, pre, specialized, workflows, modernizing, joins, www, com, related, links, cto, automate, makes, trustworthy, deeper, explicit, remain, additionally, focus, less, plausible, behaves, correctly, conditions, rejecting, slowing, down, sake, ceremony, aim, here, answer, rebalance, move, 10x, surrounding, lifecycle, adapt, documentation, incident, response, otherwise, organization, moves, choreography, operations, rest, magically, accelerate, got, prototyping, discovery, quickly, purposes, fits, past, slowed, accumulation, could, write, much, brake, disappearing, mode, expensive, reason, fragile, pressure, easy, coherence, discipline, actually, produces, machine, forces, honest, conversation, output, equal, progress, similarly, typing, contest, goal, create, understandable, continue, serving, over, humans, decide, worth, tool, violates, known, rule, there, vulnerable, feature, exist, coherent, operational, sustainable, implementation, solves, actual, lives, important, static, scanning, remediation, necessary, volume, goes, substitute, judgment, care, was, written, cares, who, breaks, paged, prompt, toolchain, team, owns, line, firm, issue, itself, always, existed, nobody, feels, fully, responsible, felt, hand, ambiguity, talk, these, like, collaborators, day, way, autocomplete, text, message, removes, friction, remove, think, matters, rarely, clean, places, edges, load, permissions, partial, two, correct, make, incompatible, each, other, becomes, blast, radius, weak, grows, introduce, interaction, another, place, break, exploitable, broken, convincing, isolation, then, meets, integrations, large, giving, which, lead, some, problems, pleasers, excellent, syntax, compiles, runs, validity, confidence, earned, dramatically, this, going, any, soon, thing, apmdigest, features, home, menu, skip, content,

Text of the page (random words):
ai generated code isn t the risk untested code is devopsdigest skip to main content devopsdigest main menu home features blog forum news events reports papers webinars tools apmdigest ai generated code isn t the risk untested code is june 08 2026 johnny halife southworks ai is making code dramatically cheaper to produce this innovation is very useful for developers and it will not be going away any time soon however cheaper code is not the same thing as better software large language models are very good at giving people what they ask for which can also lead to some problems they are people pleasers with excellent syntax and they can produce code that compiles runs and looks right in a pull request that surface level validity creates confidence before the system has earned it the risk is not that ai writes broken code the risk is that it writes convincing code code that looks good in isolation passes the obvious checks but then fails when it meets real users real data real scale and real integrations the fast and the exploitable as ai generated code becomes cheaper the blast radius of weak engineering grows teams can generate more code ship more changes and introduce more interactions between systems but every new interaction is another place where assumptions can break ai removes friction from writing code but it does not remove the need to think through failure that matters because software rarely fails in the clean obvious places it fails at the edges during integration under load across permissions with partial data during deployment or when two correct systems make incompatible assumptions about each other the faster code is generated the more deliberate the validation has to become the ownership problem ai generated code also creates ambiguity around ownership teams talk about these tools like collaborators but they are still tools at the end of the day in the same way that autocomplete does not own your text message a model does not own your production system the issue is not bad code by itself because bad code has always existed shipping systems that nobody feels fully responsible for because the work felt automated is the real problem at hand production does not care how code was written it only cares who fixes it when it breaks when a system fails someone still gets paged not the model or the prompt or the toolchain a human team owns the outcome and that line needs to stay firm automation has limits automated review tools will become more important as code velocity increases static analysis dependency scanning policy checks test generation and automated remediation are all necessary guardrails when volume goes up however guardrails are not a substitute for judgment a tool can tell you whether code violates a known rule whether a test passes and whether there s a vulnerable dependency it cannot tell you whether the feature should exist whether the architecture is coherent whether the operational model is sustainable or whether the implementation solves the actual business problem that gap is where enterprise software lives ai can generate implementations but humans still need to decide whether those implementations are worth shipping what productivity means now ai forces a more honest conversation about productivity more output does not automatically equal more progress similarly software development is not a typing contest the goal is to create systems that work stay understandable and continue serving the business over time without discipline ai does not eliminate technical debt instead it actually produces technical debt at machine speed in the past human limits slowed that accumulation engineers could only write review and ship so much code but that natural brake is disappearing the new failure mode is fast delivery of systems that become expensive to change hard to reason about and fragile under pressure speed is easy but coherence is hard where vibe coding fits vibe coding is useful for experimentation prototyping product discovery and exploring interactions quickly and teams should use it for those purposes however experimentation is not production production software is a choreography across product design engineering operations security and the business code is only one part of that system the rest does not magically accelerate just because code generation got faster if development velocity increases by 10x the surrounding lifecycle has to adapt testing review deployment monitoring documentation incident response and ownership otherwise the organization just moves risk downstream faster move fast and ask better questions rejecting ai generated code or slowing teams down for the sake of ceremony isn t the aim here the answer is to rebalance the system around the new speed testing has to go deeper ownership has to stay explicit and architecture has to remain deliberate additionally the review process needs to focus less on whether the code looks plausible and more on whether the system behaves correctly under real conditions generated code is not proven code it is a starting point ai changes the cost of source code but it does not change what makes software trustworthy someone still has to ask the hard questions validate the assumptions and own the outcome that is the part we cannot automate away johnny halife is cto and partner at southworks related links www southworks com southworks joins the vendor forum industry news ibm bob updated with multi agent capabilities july 09 2026 ibm announced major updates to ibm bob its agentic software development platform including new multi agent capabilities built in ai cost and use analytics and pre built specialized workflows for modernizing enterprise systems rapidfort and reversinglabs partner on open source dependency libraries july 09 2026 rapidfort and reversinglabs rl announced a strategic partnership to deliver rapidfort open source dependency libraries an open source package catalog to combine rapidfort s proven curation and hardening process with independent third party validation powered by reversinglabs spectra assure platform ibm red hat and deloitte announce lightwell collaboration july 08 2026 deloitte ibm and red hat announced a collaboration to help protect the software supply chain against increasingly automated cyber threats unisys and antenna partner to deliver insights for ai assisted software development july 08 2026 unisys announced a strategic partnership with antenna to embed independent third party benchmarks directly into unisys applications solutions and services entire launches distributed git network july 08 2026 entire launched the preview of its distributed git network letting developers host repositories around the world instead of on a single central provider couchbase releases ai data plane july 08 2026 couchbase announced general availability of the ai data plane a unified data infrastructure layer for enterprise ai agents perforce intelligence updated july 07 2026 perforce software announced major updates to perforce intelligence with an mcp agnostic agentic gateway an ai assisted testing and natural language execution platform and a unified compliance platform cequence platform 9 0 released july 07 2026 cequence security announced general availability of cequence platform 9 0 an ai native release that fundamentally changes how users interact with api security tools harness releases autonomous worker agents july 07 2026 harness launched autonomous worker agents for software delivery a platform for enterprises to build and safely run ai agents that handle the work between writing code and shipping it to production snaplogic releases mcp builder july 07 2026 snaplogic announced the general availability of snaplogic mcp builder a new template based capability that helps organizations operationalize ai faster by automatically turning existing integration pipelines into agent ready mcp tools baz planner released july 07 2026 baz introduced baz planner a gateway that sits between developers and the codebase automatically routing every idea through dynamic loops that instantly detect root cause patch and validate vulnerabilities and bugs proactively rewriting coding plans to eliminate entire classes of issues before they reach production gitlab recognized as a leader in the gartner magic quadrant for devsecops platforms for the fourth consecutive year june 25 2026 gitlab has been named a leader in the 2026 gartner magic quadrant for devsecops platforms sembi releases testrail 10 5 and xray 15 0 june 25 2026 sembi announced ai test prioritization in testrail 10 5 and xray 15 0 azul announces free jvm vulnerability risk assessment june 25 2026 azul launched a free jvm vulnerability risk assessment to address the blind spot that autonomous ai exploitation tools are increasingly able to find cdata launches connect ai developer edition python sdk and cli june 24 2026 cdata software launched three products for developers building ai applications on enterprise data connect ai developer edition free the cdata connect ai python sdk open source and cdata cli more industry news email signup sign up for devopsdigest email search form upcoming webinars delivering amazing digital experiences with gitlab ci july 14 2026 c c continuous testing july 15 2026 c c software testing august 19 2026 on demand webinars ai generated code for critical systems can we trust it how to test microservices without waiting on downstream services bridging googletest unit testing compliance in safety critical software gitlab transcend virtual event check point sase internet access optimization performance tuning check point sase identity integration access policy design best practices smarter pull requests ai that tests fixes validates java code improving embedded software quality with clion parasoft c c test generate api mocks autonomously in your build pipeline with your llm test impact analysis explained escaping regression slowdowns in long lived software building realistic service dependencies with your llm client how to achieve cra compliance starting with secure by design appsec at scale actionable insights from 1 000 cyber range events gitlab transcend exploring the true potential of agentic ai for software delivery all webinars analyst reports certified by design making googletest ready for safety critical software development 2026 gartner magic quadrant for devsecops platforms miercom 2026 hybrid mesh network security benchmark report frost sullivan 2026 best practices recognition global web application firewall waf and api security industry gigaom application api security radar 2026 gigaom cloud network security radar 2025 2025 gartner magic quadrant for email security 2025 gartner magic quadrant for ai code assistants 2025 gartner magic quadrant for hybrid mesh firewall 2025 gigaom radar for anti phishing the forrester wave zero trust platforms q3 2025 miercom 2025 enterprise and hybrid mesh firewall security report gigaom 2025 attack surface management radar all analyst reports white papers check point illumio hybrid mesh security in the age of ai 2026 cloud security report securing the ai transformation application security at scale insights from 1 000 cyber range events the intelligent software development era cyber security report 2026 googletest adoption challenges for safety critical code guide to api security how to maximize functional testing productivity with ai and a lean web ui test strategy performance testing best practices guide the essential guide to environment based testing the economics of software innovation waf comparison project a real world comparison cloud security report 2025 ai security report 2025 the state of cyber security 2025 all white papers media partners global rss feed about navigating devopsdigest sponsors sponsor program editorial guidelines vendor forum rules sponsor blog guidelines pr tips editor blog contact user login username password request new password the latest why devops teams build genai tooling while most organizations still rely on manual monitoring july 09 2026 agentic development and ai sdlc closing the ai coding velocity gap july 08 2026 the data quality crisis undermining enterprise analytics july 07 2026 testing at the speed of ai how agentic service virtualization helps teams validate faster 2 june 26 2026 testing at the speed of ai how agentic service virtualization helps teams validate faster 1 june 25 2026 the hidden cost of misconfigurations in hybrid cloud june 24 2026 your next customer isn t a person it s their agent june 23 2026 beyond autoscaling why smart vm scheduling is the next cost frontier for devops june 22 2026 ai has outpaced how engineering organizations measure developer productivity june 18 2026 beyond the gateways api gateways miss real development risks june 17 2026 what tighter visa scrutiny means for engineers building an extraordinary ability case june 16 2026 governing software delivery in the age of code abundance june 15 2026 agentic ai is accelerating how software gets built and how it gets attacked most enterprises are only ready for one june 12 2026 ai is rewriting what it means to be a great developer at every experience level june 11 2026 why devops teams are feeling the ai talent crunch june 10 2026 ai in sdlc right now what s working and what isn t june 09 2026 ai generated code isn t the risk untested code is june 08 2026 gartner enterprise ai coding agent entering new phase of expansion and competitive realignment june 05 2026 why itops need right sized ai not bigger models june 05 2026 the end of reactive devops ai driven observability for zero defect digital systems june 05 2026 hot topics agile ai powered dev ai ml aiops analytics api integration apm automation bizdevops ci cd cloud cloud native containers k8s database dataops dev culture development coding devex devops devsecops digital transformation e commerce iac low code no code mainframe mlops mobile monitoring open source platform engineering remote work serverless testing quality tokenmaxxing vibe coding vsm