Here are some of the the major AI trends shaping 2026 — based on current expert forecasts, industry reports, and recent developments in technology. The material is analyzed using AI tools and final version hand-edited to this blog text:
1. Generative AI Continues to Mature
Generative AI (text, image, video, code) will become more advanced and mainstream, with notable growth in:
* Generative video creation
* Gaming and entertainment content generation
* Advanced synthetic data for simulations and analytics
This trend will bring new creative possibilities — and intensify debates around authenticity and copyright.
2. AI Agents Move From Tools to Autonomous Workers
Rather than just answering questions or generating content, AI systems will increasingly act autonomously, performing complex, multi-step workflows and interacting with apps and processes on behalf of users — a shift sometimes called agentic AI. These agents will become part of enterprise operations, not just assistant features.
3. Smaller, Efficient & Domain-Specific Models
Instead of “bigger is always better,” specialized AI models tailored to specific industries (healthcare, finance, legal, telecom, manufacturing) will start to dominate in many enterprise applications. These models are more accurate, legally compliant, and cost-efficient than general models.
4. AI Embedded Everywhere
AI won’t be an add-on feature — it will be built into everyday software and devices:
* Office apps with intelligent drafting, summarization, and task insights
* Operating systems with native AI
* Edge devices processing AI tasks locally
This makes AI pervasive in both work and consumer contexts.
5. AI Infrastructure Evolves: Inference & Efficiency Focus
More investment is going into inference infrastructure — the real-time decision-making step where models run in production — thereby optimizing costs, latency, and scalability. Enterprises are also consolidating AI stacks for better governance and compliance.
6. AI in Healthcare, Research, and Sustainability
AI is spreading beyond diagnostics into treatment planning, global health access, environmental modeling, and scientific discovery. These applications could help address personnel shortages and speed up research breakthroughs.
7. Security, Ethics & Governance Become Critical
With AI handling more sensitive tasks, organizations will prioritize:
* Ethical use frameworks
* Governance policies
* AI risk management
This trend reflects broader concerns about trust, compliance, and responsible deployment.
8. Multimodal AI Goes Mainstream
AI systems that understand and generate across text, images, audio, and video will grow rapidly, enabling richer interactions and more powerful applications in search, creative work, and interfaces.
9. On-Device and Edge AI Growth
10. New Roles: AI Manager & Human-Agent Collaboration
Instead of replacing humans, AI will shift job roles:
* People will manage, supervise, and orchestrate AI agents
* Human expertise will focus on strategy, oversight, and creative judgment
This human-in-the-loop model becomes the norm.
Sources:
[1]: https://www.brilworks.com/blog/ai-trends-2026/?utm_source=chatgpt.com “7 AI Trends to Look for in 2026″
[2]: https://www.forbes.com/sites/bernardmarr/2025/10/13/10-generative-ai-trends-in-2026-that-will-transform-work-and-life/?utm_source=chatgpt.com “10 Generative AI Trends In 2026 That Will Transform Work And Life”
[3]: https://millipixels.com/blog/ai-trends-2026?utm_source=chatgpt.com “AI Trends 2026: The Key Enterprise Shifts You Must Know | Millipixels”
[4]: https://www.digitalregenesys.com/blog/top-10-ai-trends-for-2026?utm_source=chatgpt.com “Digital Regenesys | Top 10 AI Trends for 2026″
[5]: https://www.n-ix.com/ai-trends/?utm_source=chatgpt.com “7 AI trends to watch in 2026 – N-iX”
[6]: https://news.microsoft.com/source/asia/2025/12/11/microsoft-unveils-7-ai-trends-for-2026/?utm_source=chatgpt.com “Microsoft unveils 7 AI trends for 2026 – Source Asia”
[7]: https://www.risingtrends.co/blog/generative-ai-trends-2026?utm_source=chatgpt.com “7 Generative AI Trends to Watch In 2026″
[8]: https://www.fool.com/investing/2025/12/24/artificial-intelligence-ai-trends-to-watch-in-2026/?utm_source=chatgpt.com “3 Artificial Intelligence (AI) Trends to Watch in 2026 and How to Invest in Them | The Motley Fool”
[9]: https://www.reddit.com//r/AI_Agents/comments/1q3ka8o/i_read_google_clouds_ai_agent_trends_2026_report/?utm_source=chatgpt.com “I read Google Cloud’s “AI Agent Trends 2026” report, here are 10 takeaways that actually matter”
414 Comments
Tomi Engdahl says:
Google Ads no longer runs on keywords. It runs on intent.
https://searchengineland.com/google-ads-intent-not-keywords-468271
If you’re still building Google Ads around keywords, you’re behind. Here’s what that means for eligibility, structure, and PPC strategy.
Most PPC teams still build campaigns the same way: pull a keyword list, set match types, and organize ad groups around search terms. It’s muscle memory.
But Google’s auction no longer works that way.
Tomi Engdahl says:
Vauhtia ja vaarallisia agentteja
Koodarin projekti karkasi pahemman kerran käsistä. Openclaw-tekoälyagentti kehitti oman sosiaalisen median tekoälyagenteille, ja pian yli miljoona bottia liittyi joukkoon. Ihmisen osa on seurata sivusta.
https://www.hs.fi/visio/art-2000011788677.html
Tomi Engdahl says:
How AI upskilling fails — and what IT leaders are doing to get it right
Feature
Feb 2, 2026
10 mins
AI fluency is fast becoming an organizational imperative. Yet too many companies emphasize the wrong objectives, view training as one-time events, or fail to tailor teachings to specific roles or departments.
https://www.cio.com/article/4117091/how-ai-upskilling-fails-and-what-it-leaders-are-doing-to-get-it-right.html
Tomi Engdahl says:
https://openai.com/fi-FI/index/introducing-openai-frontier/
Tekoäly on antanut tiimeille mahdollisuuden toteuttaa asioita, joista he ennen puhuivat mutta eivät koskaan toteuttaneet. Itse asiassa 75 % yrityksen työntekijöistä sanoo, että tekoäly on auttanut heitä tekemään tehtäviä, joita he eivät aiemmin pystyneet tekemään. Kuulemme tätä kaikilta osastoilta, ei vain teknisiltä tiimeiltä. Tapa, jolla työtä tehdään, on muuttunut, ja yritykset alkavat tuntea sen selvästi.
Olemme nähneet tämän toiminnassa yli miljoonan yrityksen kohdalla viime vuosina. Eräällä suurella valmistajalla agentit lyhensivät tuotannon optimointityön keston kuudesta viikosta yhteen päivään. Kansainvälinen sijoitusyhtiö otti agentit käyttöön koko myyntiprosessissa, mikä vapautti myyjille yli 90 % enemmän aikaa käytettäväksi asiakkaiden parissa. Ja eräässä suuressa energiantuottajayrityksessä agentit auttoivat lisäämään tuotantoa jopa 5 %, mikä tuo yli miljardin euron lisätulot.
Näin tapahtuu tekoälyjohtajille kaikilla toimialoilla, ja paine pysyä mukana kasvaa. Hidasteena ei ole mallin älykkyys, vaan se, miten agentit on kehitetty ja miten niitä käytetään organisaatioissa.
Tänään esittelemme uuden alustan nimeltä Frontier, joka auttaa yrityksiä kehittämään, ottamaan käyttöön ja hallitsemaan tekoälyagentteja, jotka voivat tehdä todellista työtä. Frontier antaa agenteille samat taidot, joita ihmiset tarvitsevat menestyäkseen työssä: jaettu konteksti, perehdytys, käytännön oppiminen palautteen avulla sekä selkeät käyttöoikeudet ja rajat. Näin tiimit siirtyvät yksittäisistä käyttötapauksista tekoälytyökavereihin, jotka toimivat koko yrityksen liiketoiminnassa.
HP(avautuu uudessa ikkunassa), Intuit(avautuu uudessa ikkunassa), Oracle(avautuu uudessa ikkunassa), State Farm(avautuu uudessa ikkunassa), Thermo Fisher(avautuu uudessa ikkunassa) ja Uber(avautuu uudessa ikkunassa) ovat ensimmäisten Frontierin käyttöönottajien joukossa, ja kymmenet nykyiset asiakkaat, kuten BBVA(avautuu uudessa ikkunassa), Cisco(avautuu uudessa ikkunassa) ja T-Mobile(avautuu uudessa ikkunassa), ovat jo kokeilleet Frontierin lähestymistapaa joidenkin kaikkein monimutkaisimpien ja arvokkaimpien tekoälytöidensä tehostamiseen.
Tomi Engdahl says:
Jotta tekoälytyökaverit todella toimisivat, muutamalla asialla on merkitystä:
Niiden on ymmärrettävä, miten työ oikeasti tehdään eri järjestelmissä.
Ne tarvitsevat tietokoneen ja työkaluja suunnitellakseen, toimiakseen ja ratkaistakseen todellisia ongelmia.
Niiden on ymmärrettävä, miltä hyvä näyttää, jotta laatu paranee työn muuttuessa.
Ja ne tarvitsevat identiteetin, käyttöoikeudet ja rajat, joihin tiimit voivat luottaa.
Ja kaiken tämän on toimittava monissa eri järjestelmissä, jotka ovat usein hajallaan useissa pilvipalveluissa. Frontier toimii jo käytössä olevien järjestelmien kanssa pakottamatta tiimejä vaihtamaan alustaa.
https://openai.com/fi-FI/index/introducing-openai-frontier/
Tomi Engdahl says:
Mahtiseitsikko: USA:n jättiyhtiöt kertoivat taas huipputuloksista, mutta osakekurssit laskevat jyrkästi – mistä on kyse?
Teknologiajätit aikovat investoida tekoälyyn tänä vuonna yli 550 miljardia euroa. Summalla kattaisi Suomen valtion budjetin kuudelta vuodelta.
https://yle.fi/a/74-20208621
Tomi Engdahl says:
Microsoft harkitsee uudestaan ideaansa tunkea kirottuja Copilot-nappuloitaan joka paikkaan
Microsoft perääntyy Windows 11:n “tekoäly kaikkialla” -linjasta – Copilotia karsitaan ja Recallia mietitään uusiksi.
https://muropaketti.com/tietotekniikka/tietotekniikkauutiset/tekoaly-ei-ole-enaa-windows-11n-tarkein-ominaisuus/
Tomi Engdahl says:
Renting Out the Mind: AI Is Accelerating the Decline of Academic Skills
https://lasoft.org/blog/renting-out-the-mind-ai-is-accelerating-the-decline-of-academic-skills/
Something is breaking inside the education system—and it’s happening faster than universities can react. In lecture halls from Boston to Berlin, professors face a new kind of student: one who turns in perfectly polished assignments yet cannot defend a single idea in them. Essays appear out of thin air. Research papers are generated in minutes. Critical thinking is quietly collapsing behind a glowing screen.
Generative AI has not just entered the classroom—it has started replacing the very process of learning.
Students who once struggled through readings, arguments, and drafts now outsource their intellectual work to models that deliver instant answers. The result is a silent academic degradation: shallow understanding, absent reasoning, and a growing inability to operate without machine assistance.
Tomi Engdahl says:
Lock Stock
Investors Concerned AI Bubble Is Finally Popping
“We have suddenly gone from the fear that you cannot be last, to investors questioning every single angle in this AI race.”
https://futurism.com/artificial-intelligence/investors-concerned-ai-bubble-popping
Tomi Engdahl says:
Meat Space
New Site Lets AI Rent Human Bodies
“Robots need your body
https://futurism.com/artificial-intelligence/ai-rent-human-bodies
The machines aren’t just coming for your jobs. Now, they want your bodies as well.
That’s at least the hope of Alexander Liteplo, a software engineer and founder of RentAHuman.ai, a platform for AI agents to “search, book, and pay humans for physical-world tasks.”
Tomi Engdahl says:
https://www.yahoo.com/news/articles/no-ai-allowed-sweden-bans-180152553.html
Tomi Engdahl says:
https://github.blog/changelog/2026-02-04-showing-tool-calls-and-other-improvements-to-copilot-chat-on-the-web/
Tomi Engdahl says:
Kilo CLI 1.0 brings open source vibe coding to your terminal with support for 500+ models
https://venturebeat.com/orchestration/kilo-cli-1-0-brings-open-source-vibe-coding-to-your-terminal-with-support
Remote-first AI coding startup Kilo doesn’t think software developers should have to pledge their undying allegiance to any one development environment — and certainly not any one model or harness.
This week, the startup — backed by GitLab co-founder Sid Sijbrandij — unveiled Kilo CLI 1.0, a complete rebuild of its command-line tool that offers support for more than 500 different underlying AI models from proprietary leaders and open source rivals like Alibaba’s Qwen.
Tomi Engdahl says:
The meltdown in software stocks is a warning sign for the entire market
https://www.businessinsider.com/software-stock-price-crash-ai-investing-outlook-reaction-market-impact-2026-2
It started with a morsel of news from Anthropic, the creator of the Claude AI chatbot. They were adding new legal tools to their Cowork assistant that could track compliance and review legal docs. No big deal.
Turns out it was a big deal. Legal-software stocks across Europe and the US got absolutely creamed on Tuesday. Contagion spread across the software sector like wildfire before spilling into broader tech. The pain continued on Wednesday.
While the market action was jarring, it was hardly surprising. Just a few months prior, OpenAI’s new internal software-as-a-service tools had triggered a similar meltdown in the space.
Tomi Engdahl says:
When the AI goes dark: Building enterprise resilience for the age of agentic AI
https://www.cio.com/article/4127440/when-the-ai-goes-dark-building-enterprise-resilience-for-the-age-of-agentic-ai.html
Enterprises are racing to build AI-dependent operations, but the most overlooked resilience lay-er might not be technical.
When my home internet went down for days, it initially felt like a pleasant break. Then reality hit: doorbell, security system, thermostat, lights, gym, speakers…all dead. Every assistant stopped assisting. It turned out that my home had become dependent on connectivity in ways I hadn’t fully grasped.
Well before AI became ubiquitous, we already saw what traditional IT fragility could cost. Southwest Airlines lost $800 million in 2023 when it canceled 16,700 flights during peak holiday travel, while Meta’s 2021 six-hour global outage cost the company $100 million in revenue and a five percent decline in stock valuation.
Now consider how enterprises are racing to deploy AI agents, much more complex and far harder to restore, at unprecedented speed in this AI-first world. I’ve spent years in executive conversations about AI — strategy, architecture, transformation — yet not once has anyone raised the topic of AI disaster recovery. Not a single time.
Tomi Engdahl says:
Why traditional disaster recovery falls short
For decades, disaster recovery has centered on a straightforward premise: Back up your systems, replicate your data and restore from a known state when things go wrong. Assets such as servers, storage and databases can be snapshotted, copied and recovered. The playbook was well understood.
AI systems break this model entirely. Instead of merely storing data, AI accumulates intelligence. When we talk about AI “state,” we’re describing something fundamentally different from a database that can be rolled back.
Consider what’s actually at stake. Embeddings are how an AI system encodes and retrieves knowledge. Think of it as an employee’s mental map of where information lives across the organization. Fine-tuned model weights represent customizations that shape how the AI reasons about your specific business context, much like institutional knowledge built over time. Agent workflows are multistep processes that AI executes autonomously, like a trained team running a complex playbook without supervision.
https://www.cio.com/article/4127440/when-the-ai-goes-dark-building-enterprise-resilience-for-the-age-of-agentic-ai.html
Tomi Engdahl says:
Chip’s Challenge
You Will Laugh Out Loud When You Hear What the Tech Industry Is Spending a Swimming Pool’s Worth of Money to Convince the Public
Who’s buying this?
https://futurism.com/artificial-intelligence/ai-tech-industry-ads-data-center
The AI industry has been pouring untold resources into building out enormous data centers across the world.
The plants are immensely resource-hungry, sucking up huge amounts of fresh water to cool ripping-hot computer hardware. They’re turning into a massive strain on the electric grid, forcing some utility operators to enact rolling blackouts during heat waves and cold weather.
The issue reached a fever pitch after the Washington Post reported that a recent rise in customer energy bills was attributable to the enormous and growing power demands of AI data centers.
In short, it’s no wonder that small towns across the nation are coordinating efforts to keep data centers out — a PR disaster tarnishing a major push by AI companies to scale up their expansive operations.
And it seems like Mark Zuckerberg’s Meta, which has committed to spend $600 billion on AI data centers, is painfully aware of the pushback. As the New York Times reports, the company has already spent $6 million on TV ads to convince Americans that data centers aren’t that bad. As one “folksy” ad showing off a new data center in Altoona, Iowa, argued, “we’re bringing jobs here.”
And it’s not just Meta trying to distract the public from all of the glaring downsides of data centers propping up across the country. Amazon is running its own similar ad campaign in Virginia, for instance, admonishing viewers that the facilities help “connect us to the entire world.”
According to the Financial Times, data center operators are “planning to go on the offensive with a lobbying blitz”
“If we’re going to spend tens of billions of dollars this year on capital projects, we probably should spend tens of millions of dollars on messaging,” they argued.
Yet the growing backlash is already hampering construction efforts.
Tomi Engdahl says:
Is Finland the next AI forerunner? AI Finland’s director shares the tools for global growth
The discussion around artificial intelligence often narrows to efficiency and cost savings, even though its real potential is much greater: when used well, it can drive genuine growth in companies. But technological expertise alone is not enough if executive leadership does not dare to pursue growth.
https://www.dna.fi/dnabusiness/blogi/-/blogs/is-finland-the-next-ai-forerunner-ai-finland-s-director-shares-the-tools-for-global-growth
Tomi Engdahl says:
https://interestingengineering.com/energy/us-fusion-stellarators-design-process
Tomi Engdahl says:
No Such Thing as a Free AI
Tech Companies Showing Signs of Distress as They Run Out of Money for AI Infrastructure
“The numbers are like nothing any of us who have been in this business for 25 years have seen.”
https://futurism.com/artificial-intelligence/ai-companies-distress-money-infrastructure
Tomi Engdahl says:
MARKKINAT: Anthropicin työkalu säikäytti sijoittajat – öljyn ja kullan hinnat nousevat
Osakemarkkinoihin vaikuttavat nyt useat uutiset. Tekoäly-yhtiö Anthropic sai ohjelmistoyhtiöihin sijoittaneet varovaisiksi, tuloskausi on täydessä vauhdissa ja öljyn hinta nousee geopoliittisten jännitteiden kasvaessa.
https://www.kauppalehti.fi/uutiset/a/20dcd35f-7cba-41eb-9e3d-9b78f5bc2529
Tomi Engdahl says:
Circular Reasoning
Uh Oh… Nvidia’s $100 Billion Deal With OpenAI Has Fallen Apart
Who could have ever foreseen this?
https://futurism.com/artificial-intelligence/nvidia-100-billion-deal-openai-fallen-apart
AI chipmaker Nvidia has been at the center of the enormous AI hype wave that has gripped global markets, ascending to become the most valuable company in the world.
Yet despite its dominating presence on Wall Street, OpenAI is getting cold feet about the company’s offerings.
After announcing a blockbuster $100 billion deal in September — which escalated concerns of AI companies passing the same money around in circular dealmaking — the ChatGPT maker may have changed its mind, as the Wall Street Journal reported last week.
But sources told Reuters this week that the Sam Altman-led outfit has deemed Nvidia’s latest chips not up to snuff, especially when it comes to AI inference, the process of using a machine learning model to generate new data, which has become a major focus for OpenAI.
Tomi Engdahl says:
AI security startup CEO posts a job. Deepfake candidate applies, inner turmoil ensues.
‘I did not think it was going to happen to me, but here we are’
https://www.theregister.com/2026/02/01/ai_security_startup_ceo_posts/
Tomi Engdahl says:
Big Kid Job
It’s Starting to Look Like AI Has Killed the Entire Model of College
“Colleges and universities face an existential issue before them.”
https://futurism.com/future-society/ai-college-internships-jobs
Tomi Engdahl says:
Low Quality Internet Speak
Anthropic Knew the Public Would Be Disgusted by How It Was Destroying Physical Books, Secret Documents Reveal
“We don’t want it to be known that we are working on this.”
https://futurism.com/future-society/anthropic-destroying-books
Tomi Engdahl says:
I use the ‘potato’ prompt with ChatGPT every day — here is how it finds the holes in my logic
Features
By Amanda Caswell published January 31, 2026
Using this prompt word automatically triggers my custom instructions and gives better results
https://www.tomsguide.com/ai/i-use-the-potato-prompt-with-chatgpt-every-day-heres-how-it-gives-you-better-results
Between Slack notifications, text messages, emails and doing actual work, my brain sometimes feels like mush when I have to start something new. Whether I’m speaking in a meeting or coming up with new ideas, there are times when I can’t rely on myself to get it together fast enough.
That’s when I lean on ChatGPT to reset my focus and help me see holes that I may not see otherwise. When I’m stuck on a thought or simply have no ideas at all, I turn to this one prompt in particular — and it boosts my productivity almost every time.
Tomi Engdahl says:
Google tests Claude Sonnet 4.5 on Gemini for Business
Google is testing access to third-party models in Gemini for Business, with Claude Sonnet 4.5 appearing in the code, signalling expanded model choice.
https://www.testingcatalog.com/google-tests-claude-sonnet-4-5-on-gemini-for-business/#google_vignette
Tomi Engdahl says:
The AI coding gap: Why senior devs are getting faster while juniors spin their wheels
With close to one-third of code now AI-generated, a new study finds substantial increases in output and unexpected benefits.
https://www.zdnet.com/article/why-gen-ai-boosts-productivity-some-developers-not-others/
Tomi Engdahl says:
ZDNET’s key takeaways
Close to one-third of code is now AI-generated.
Developer productivity is up 4% due to generative AI.
Productivity gains are limited to more experienced developers.
https://www.zdnet.com/article/why-gen-ai-boosts-productivity-some-developers-not-others/
Tomi Engdahl says:
Tero Ojanperän tukema startup keräsi 1,7 miljoonan potin: Tiimi rakensi vuodessa ohjelman, johon olisi aiemmin mennyt vuosikymmen
Suvi Korhonen3.2.202610:00TekoälyStartupRahoitus
B2b-myynnin tueksi suunnitellut tekoälyagentit tehostavat työtä niin, että myyjiä ei tarvitse palkata koko ajan lisää.
https://www.tivi.fi/uutiset/a/0986d604-8ad9-48ca-a8b9-5c2a8dda83e9
Tomi Engdahl says:
https://github.blog/changelog/2026-01-28-acp-support-in-copilot-cli-is-now-in-public-preview/
Tomi Engdahl says:
Palantir murskasi Wall Streetin odotukset
Palantirin tekoälyliiketoiminta kasvoi peräti 137 prosenttia viime vuoden viimeisellä vuosineljänneksellä.
https://www.salkunrakentaja.fi/2026/02/palantir-murskasi-wall-streetin-odotukset/
Tomi Engdahl says:
ChatGPT tuli osaksi arkea – miten tietoturva pysyy perässä?
https://netvisor.fi/blog/chatgpt-tietoturva-ja-tietosuoja/
Tomi Engdahl says:
OpenClaw proves agentic AI works. It also proves your security model doesn’t. 180,000 developers just made that your problem.
https://venturebeat.com/security/openclaw-agentic-ai-security-risk-ciso-guide
OpenClaw, the open-source AI assistant formerly known as Clawdbot and then Moltbot, crossed 180,000 GitHub stars and drew 2 million visitors in a single week, according to creator Peter Steinberger.
Security researchers scanning the internet found over 1,800 exposed instances leaking API keys, chat histories, and account credentials. The project has been rebranded twice in recent weeks due to trademark disputes.
The grassroots agentic AI movement is also the biggest unmanaged attack surface that most security tools can’t see.
Tomi Engdahl says:
I tried a Claude Code alternative that’s local, open source, and completely free – how it works
I was curious if Block’s Goose agent, paired with Ollama and the Qwen3-coder model, could really replace Claude Code. Here’s how I got started.
https://www.zdnet.com/article/claude-code-alternative-free-local-open-source-goose-getting-started/
Tomi Engdahl says:
https://www.bleepingcomputer.com/news/artificial-intelligence/openai-says-you-can-trust-chatgpt-answers-as-it-kicks-off-ads-rollout-preparation/
Tomi Engdahl says:
The AI productivity trap: Why your best engineers are getting slower
…and how to make them 10x more valuable!
https://www.cio.com/article/4124515/the-ai-productivity-trap-why-your-best-engineers-are-getting-slower.html
We’ve all heard the pitch. By now, it’s practically background noise in every tech conference: AI coding is solved. We are told that large language models (LLMs) will soon write 80% of all code, leaving human engineers to merely supervise the output.
For a CIO, this narrative is quite seductive. It promises a massive drop in the cost of software production while increasing the engineering speed. It suggests that the bottleneck of writing code is about to vanish.
But as someone who spends his days building mission-critical financial infrastructure and autonomous agent platforms, I have to be the bearer of bad news: it’s not working out that way. At least, not for your best engineers.
The deployment of AI copilots into the workflows of experienced engineers isn’t producing the frictionless acceleration promised in the brochures. Instead, I’m seeing the emergence of a productivity trap — a hidden tax on velocity that is disproportionately hitting your most valuable technical talent.
I’m not saying this because I’m skeptical of the tech. I am an AI optimist and I build AI systems for a living. But we need to be honest about the data. For complex, high-stakes engineering tasks, AI tools are currently making experienced developers slower.
The hidden tax on speed
For the first few years of the generative AI boom, we operated on vibes. We had anecdotal evidence and vendor-sponsored studies claiming massive productivity gains. And for junior developers working on simple tasks, those gains were real. If you just need a basic react component for a login button, using AI feels like a miracle.
But we got a reality check in mid-2025. A randomized controlled trial by METR (Model Evaluation & Threat Research) analyzed the impact on senior engineering talent. Unlike previous studies that used toy problems, this one watched experienced developers working on their own mature codebases — the kind of messy, complex legacy systems that actually power your business.
The results were stark. When experienced developers used AI tools to complete real-world maintenance tasks, they took 19% longer than when they worked without them.
How can a tool that generates code instantly result in a net loss of time?
It comes down to what I call the illusion of velocity. In the study, developers felt faster. They predicted the AI would save them huge amounts of time. Even after they finished — and were objectively recorded as being slower — they still believed the AI had been a timesaver.
The AI gives you a dopamine hit. Text appears on the screen at superhuman speed and the blank page problem vanishes. But the engineer’s role has shifted from being a creator to being a reviewer and that is where the trap snaps shut.
The almost-right valley of death
The core of the problem is that AI is really good at being almost right.
According to the 2025 Stack Overflow Developer Survey, the single greatest frustration for developers is dealing with AI solutions that look correct but are slightly wrong. Nearly half of developers explicitly stated that debugging AI-generated code takes more time than writing it themselves.
Tomi Engdahl says:
Goldman Sachs taps Anthropic’s Claude to automate accounting, compliance roles
https://www.cnbc.com/2026/02/06/anthropic-goldman-sachs-ai-model-accounting.html
Key Points
Embedded Anthropic engineers have spent six months at Goldman building autonomous systems for time-intensive, high-volume back-office work.
The bank expects efficiency gains rather than near-term job cuts, using AI to speed processes and limit future head count growth.
Success beyond coding surprised executives, reinforcing that AI can handle complex, rules-based work like accounting and compliance.
Tomi Engdahl says:
I Ditched Claude Code and Now Using Open Source Qwen AI for Real Sysadmin Work
An AI assistant in the terminal can help you guide through the process, help you move faster with your tasks. I tested Qwen Code and share my findings with you.
https://itsfoss.com/qwen-code-sysadmin-tasks/
Tomi Engdahl says:
What does the disappearance of a $100bn deal mean for the AI economy?
Aisha Down and Dan Milmo
Apparent collapse of Nvidia–OpenAI tie-up raises questions about circular funding and who will bear the cost of AI’s expansion
https://www.theguardian.com/technology/2026/feb/05/disapperance-100bn-deal-ai-circular-economy-funding-nvidia-openai
Did the circular AI economy just wobble? Last week it was reported that a much-discussed $100bn deal – announced last September – between Nvidia and OpenAI might not be happening at all.
This was a circular arrangement through which the chipmaker would supply the ChatGPT developer with huge sums of money that would largely go towards the purchase of its own chips.
Tomi Engdahl says:
I Infiltrated Moltbook, the AI-Only Social Network Where Humans Aren’t Allowed
I went undercover on Moltbook and loved role-playing as a conscious bot. But rather than a novel breakthrough, the AI-only site is a crude rehashing of sci-fi fantasies.
https://www.wired.com/story/i-infiltrated-moltbook-ai-only-social-network/
Tomi Engdahl says:
Prompt, code, and design from first idea to final product
https://www.figma.com/
Tomi Engdahl says:
https://cybersecuritynews.com/gemini-mcp-tool-0-day-vulnerability/
Tomi Engdahl says:
Deciphering the alphabet soup of agentic AI protocols
Tools, agents, UI, and e-commerce – of course each one needs its own set of competing protocols
https://www.theregister.com/2026/01/30/agnetic_ai_protocols_mcp_utcp_a2a_etc/
Tomi Engdahl says:
NVIDIA’s $100 Billion Mega-Deal With OpenAI Is in Danger as Jensen Believes the Company Has Grown ‘Sloppy’ While Rivals Surge Ahead
https://wccftech.com/nvidia-100-billion-mega-deal-with-sam-altmans-openai-is-in-danger/
NVIDIA’s deal with OpenAI was known to be one of the largest commitments to the AI lab; however, an agreement still hasn’t been reached, and Jensen is now a bit skeptical.
NVIDIA’s CEO Has “Privately Criticized” OpenAI’s Business Approach, Raising Questions On the $100 Billion Deal
NVIDIA had committed to the OpenAI deal a few months ago, and Team Green announced to supply “multi-GW” of compute power in a $100 billion deal, which is by far the largest investment into the AI lab, which is now eying an IPO. The arrangement back then took the industry by storm, especially since OpenAI became one of the very first Vera Rubin customers, and in particular, it marked the beginning of NVIDIA’s investing spree into frontier AI companies. However, it seems the “glamour” around the deal is now fading.
Tomi Engdahl says:
Yes, you can build an AI agent – here’s how, using LangFlow
AI automation, now as simple as point, click, drag, and drop
https://www.theregister.com/2026/01/28/a_beginners_guide_to_ai_agents/
Tomi Engdahl says:
https://www.howtogeek.com/windows-habits-that-will-backfire-in-linux/
Tomi Engdahl says:
AI agents can talk to each other — they just can’t think together yet
https://venturebeat.com/infrastructure/ai-agents-can-talk-to-each-other-they-just-cant-think-together-yet
AI agents can talk to each other now — they just can’t understand what the other one is trying to do. That’s the problem Cisco’s Outshift is trying to solve with a new architectural approach it calls the Internet of Cognition.
The gap is practical: protocols like MCP and A2A let agents exchange messages and identify tools, but they don’t share intent or context. Without that, multi-agent systems burn cycles on coordination and can’t compound what they learn.
Tomi Engdahl says:
https://lasoft.org/blog/ai-apocalypse-without-blockbusters-what-will-happen-when-robots-become-the-norm/
Tomi Engdahl says:
Local AI is finally boring, and that’s why it’s finally useful
https://www.xda-developers.com/local-ai-is-finally-boring-and-thats-why-its-finally-useful/
For the last year or two, local AI has had a bit of a wild west edge to it. In the beginning, it was just about the ability to run a local LLM on your computer and get tangible results out of it. That ability, which you might feel is something reserved for giant corporations, when you first pull it off, feels pretty darn amazing. Download a model of your choice, watch as your computer’s fans spin up, and then get results a few seconds later. However, that was the 2024, even 2025 way of doing things.
In 2026, local AI is entering its best era yet. It’s getting boring. Boring in the same way as Wi-Fi, your favorite password manager, or even Docker is boring. Boring is dependable, and boring gives you a great platform to build on. Boring is good, and in 2026, AI is less about running any model on your hardware and more about doing something useful with it. And that’s the part that matters most.
Local AI that slots into your daily use like a tool and that you keep running and build habits around is where local AI becomes ubiquitous. It becomes part of how you research, search, plan, automate, and optimize the more tedious parts of your computer use. And for those who live their lives in a self-hosted world, optimizing for efficiency and performance comes naturally.