Pat Bertini

Early Adopter Programme Sign-up Be first. Be ready. Be AI Act compliant. The EU AI Act enters enforcement in August 2026, and only 9% of companies are prepared. AI systems cannot simply be retrofitted for compliance. Transparency, ...

The AI Act cannot be retrofitted.

If your AI is already in production, the clock is already ticking.

Euler’s Early Adopter Programme starts in April.
We embed EU compliance directly into product workflows.

Join here → forms.gle/E4c5jEhEbGuT...

#AIAct

Join a live conversation on Human In The Loop here forum.helio.app/t/what-does-...

Meta just bought Manus, an AI startup everyone has been talking about | TechCrunch Meta says it'll keep Manus running independently while weaving its agents into Facebook, Instagram, and WhatsApp, where Meta's own chatbot, Meta AI, is already available to users.

I loved @manusaihq.bsky.social - i believe it is the best agent out there.
Now, learning that Meta just acquired it leaves me AI-orphan: Llama is a biased model, and the lack of ethics in Meta means i am going to delete my subscription and look for a Manus replacement.
techcrunch.com/2025/12/29/m...

Velocity over speed: why the AI race has already failed We are confusing motion with progress. In physics, this mistake has a name: speed without direction. In civilisation, it has a cost.

Everyone’s talking about the “AI race.”
But where exactly are we racing to?
Speed without direction isn’t progress, it’s motion in circles.
I wrote about why the AI race has already failed, and why velocity matters more than speed. medium.com/@pat-bertini...
#AI #innovation

ChatGPT ads are coming, OpenAI leak suggests CEO Sam Altman once described ads as a ‘last resort’

If ChatGPT recommended a hotel, would you know if it was genuine or sponsored?
Code in the beta reveals ads are coming. Without transparency and explainability, we can't tell the difference.
This is why Responsible AI isn't optional anymore
#AiAct
www.independent.co.uk/tech/chatgpt...

Figma sued for allegedly misusing customer data for AI training Design software company Figma was hit with a proposed class action in California federal court on Friday for allegedly misusing its customers' designs to train artificial intelligence models.

Heads-up to all Figma users!
Figma is reportedly sued for allegedly using customers’ designs for AI training without clear consent. If true, it’s a serious reminder: before adding AI-driven features, ensure data use is opt-in, transparent, and clients are informed.
www.reuters.com/legal/govern...

Stop saying AI hallucinates - it doesn't. And the mischaracterization is dangerous The language we use to describe AI is pivotal to how humans see AI - so we'd better get it right.

AI isn’t hallucinating. It’s guessing. And the more we humanise failures, the less seriously organisations take risk, bias, and accountability. This article nails it:
www.zdnet.com/article/stop...

#AI

Rosenverse: Designing Within the Lines: How the EU AI Act Can Spark Better AI Innovation Are you building or using AI in your products? Do you have customers in the EU, or plan to? If so, the EU AI Act isn’t just another piece of legislati...

Everyone says regulation kills creativity. I disagree. The EU AI Act is forcing the industry to grow up.
In 2 weeks I’ll share why at Rosenfeld Media’s free event. If you design or build AI, you should be there: rosenverse.rosenfeldmedia.com/videos/desig...
#AI #responsibleAI #innovation

So, I am trying to quantify which percentage of existing AI systems (high-rish and medium-risk) can be conveniently retrofitted.
For many legacy systems, the cost of retrofitting would exceed the cost of replacement. And yet, none is taking the AI Act requirements seriously enough!

The EU’s Digital Omnibus on AI: what the proposed changes mean for your organisation Note: These are proposed changes, not yet approved law. The consultation period remains open.

The EU’s Digital Omnibus delays high-risk AI rules by up to 16 months. Simplification on paper but greater risk in practice. @the-euler.bsky.social new analysis explains what’s changing and why organisations should prepare now, not later.

#AI #AiAct

www.linkedin.com/pulse/eus-di...

The EU’s Digital Omnibus revisits multiple digital regulations and delays key AI Act rules to Dec 2027.
They call it simplification but it looks like a gift to Big Tech. Narrower personal-data rules and delayed enforcement weaken protection. We don’t need time, we need integrity.

The salt in the AI cake: seven emerging jobs no one is preparing for In the quiet corners of our technological revolution, there is a crisis no one is naming. We are building systems designed to augment human…

The Future of AI Governance is Human!

In my last article I explore 7 emerging roles essential for responsible AI: from AI Decision Auditors to Transparency Designers. We must invest in these positions to ensure AI serves human values, not undermine them.
pat-bertini.medium.com/the-salt-in-...

When Deepware meets Wetware: the uncomfortable truth about responsible AI There is a memo from IBM dated 1979 that should be mandatory reading for every CEO racing to deploy AI. It states simply and profoundly…

AI doesn’t fail, we do.
Deepware (AI) is only as responsible as our wetware (humans).
Accountability, oversight, ethics: these aren’t optional, they’re the real frontier.

Read 👉 pat-bertini.medium.com/when-deepwar...

#ResponsibleAI #Ethics

If you are a disk product leader, are you rethinking your workflow to embed #gdpr, #aiact, #a11y compliance into your products?

Take this survey by @aligned-consulting.bsky.social and share your experience.

Exploring DesignOps as a Business Strategic Function Exploring DesignOps as a Business Strategic Function - Download as a PDF or view online for free

Is #DesignOps a Business Strategic Function?
To answer this question, I explored the job markets in UK, US, and EU.
The learnings?
With unclear career paths and expectations, the market is left confused.
the solution? #Awareness and #Intentionality.

The Slides: www.slideshare.net/slideshow/ex...

Photo of Patrizia Bertini and the title of her ConveyUX 2025 presentation, Exploring Designops as a Business Strategic Function, scheduled February 25, 2025.

DesignOps is more than just design—it’s a strategic business driver. 🚀 Join Patrizia Bertini at #ConveyUX as she explores how DesignOps can drive innovation, efficiency, and real business impact. 👉 conveyux.com/sessions/exp...

Bias baked in Setting ‘harmonised standards’ for artificial intelligence is a crucial element of the EU’s AI Act. But who develops these standards? CEO reveals how Europe’s standard-setting bodies dealing with AI a...

A must-read to understand the risks of standardising AI risks: corporateeurope.org/en/2025/01/b...

The Ethics of Artificial Intelligence: What We Need to Consider Before AI Becomes More Powerful Than Us As AI continues to evolve and become more powerful, it raises important ethical questions that we must address. Read more about this at InnoVirtuoso.com

🤖💡 Dive into the ethical dilemmas of AI with our latest blog post: "The Ethics of Artificial Intelligence: What We Need to Consider Before AI Becomes More Powerful Than Us." Let's think ahead! 👉 innovirtuoso.com/technology/t...

#AI #Ethics #FutureTech

cannot agree more!

CDT Europe's AI Bulletin: November 2024 Consultation on Applying the AI Act’s AI System Definition and Prohibited AI Systems This month, the AI Office launched a combined multistakeholder consultation on how to apply the AI Act’s definition...

🚨 The November AI Bulletin is here!

🔵 AI Office launches consultation on applying the #AIAct's definition of AI systems & prohibited practices.

🔵 Draft General-Purpose AI Code of Practice: the first draft is out for feedback.

… and more on #AI in the EU!

📖 Read now: cdt.org/insights/cdt...

Currently, 77 policies regulating #AI are in effect, but none address healthcare. Additionally, 177 policies are under discussion. This surge in regulatory activity will significantly transform the legal landscape and requirements for digital products.

Data source: www.fairly.ai/blog/map-of-...

PAI’s Responsible Practices for Synthetic Media PAI's Responsible Practices for Synthetic Media is a framework on how to responsibly develop, create, and share synthetic media

synthetic media, created/modified w/ #AI is raising #ethics questions.
#PAI developed a framework for responsible creation and distribution focusing on transparency, disclosure, informed consent, labelling, detection and response protocols.
syntheticmedia.partnershiponai.org#read_the_fra...

What are AI guardrails? AI guardrails help ensure that an organization’s AI tools, and their application in the business, reflect the organization’s standards, policies, and values.

Good read on #AI guardrails from McKinsey and Co. www.mckinsey.com/featured-ins...

Grok using X's real-time data to train its model risks perpetuating systemic discrimination.
Using low-quality, biased data to train a LLM violates ethical principles and laws: GDPR & AI Act both emphasise fairness, accountability, and transparency in automated decision-making.
#AI #AiEthics

I'm never sure if my cat is plotting world domination or just judging my life choices ...

#caturday... because sometimes there is more than AI and designops

I smile thinking of you smiling on a tram 🚊 😆
Going meta, i Need a coffee

I agree that service design is key to improve services by aligning people, processes, communication. However, imho while service design focuses on improving service interactions, design thinking is a strategic approach to tackle complex problems by deeply understanding users and driving innovation.

Either Freudian or i need new reading ( and writing) glasses 🤓

If you ask any #GenAI this question:
"based on our interactions, what do you know about me?"
these are what you get from ChatGPT, Perplexity, Mistral, Phind, and Claude.
ChatGPT is clearly the creepiest - time to deactivate the OpenAI's "memory" - as Phind suggest.

#AI #privacy