Sign in. It’s quick, free and it’s up to you.
An account is an optional way to support the work we do. Find out more.
THE UNITED STATES and the United Kingdom have refused to sign a global declaration on artificial intelligence at France’s global action summit on the emerging technology.
The declaration urged that the fast-moving new technology must be “open”, “ethical” and flanked by international governance.
Ireland is among 61 signatories, who also include China, France, Germany and India, who also agreed it is a priority to ensure “AI is open, inclusive, transparent, ethical, safe, secure and trustworthy, taking into account international frameworks for all.”
A UK government spokesperson cited “national interest” for Britain not signing the final statement.
It comes as US vice president JD Vance advocated for light touch regulation of AI, arguing that “excessive” regulation could “kill” the rapidly developing sector just as it is taking off.
He made the remarks during a speech this morning at the summit.
World leaders and tech heavyweights were in Paris for the second day of the summit, which aims to highlight the benefits of AI and find a common ground on the rapidly developing technology.
The Summit has been described as a first of its kind gathering, with some likening it to an AI version of Climate COPs.
The European Union, United States, China and India each have different priorities in tech development and AI regulation, and this has been visible at the conference.
France had hoped that governments would agree on voluntary commitments to make AI sustainable and environmentally friendly, but this hope was dashed by the US and UK decision not to sign the final statement.
Speaking this morning, Vance singled out the European Union’s Digital Services Act by name for criticism over “massive regulations it created about taking down content and policing so-called misinformation” that he said placed an unfair burden on American tech giants.
The US Vice President also took a swipe at China as one of several “authoritarian regimes” he said were looking to use AI for increased control of citizens at home and other countries abroad.
Taoiseach Micheál Martin was also in Paris for the conference.
Speaking to reporters this morning, the Taoiseach said the key takeaway from the conference is the “profound” implications the technology will have on people’s lives.
Asked about the rush to deregulate the emerging industry, the Taoiseach said it is fair to say there is a “rebalancing” of the approach to AI now taking place in Europe.
“That said, there were very strong voices last evening about the importance of protecting children and young people, for example, mental health, protecting democracy, hate speech and so on,” he said.
Last night, world leaders gathered for a working dinner hosted by French President Emmanuel Macron.
In a speech at the summit yesterday, Macron went toe-to-toe with the US, pledging €109bn for investment in French AI over the next five years in a bid to keep Europe in the running for the tech’s hoped-for benefits.
Macron said this was the equivalent investment for France that ChatGPT maker OpenAI has committed to investing in the US with its €500bn Stargate Project.
Announced at the end of January, the Stargate Project aims to “re-industrialise” the United States and represents a huge investment in the technology.
Macron highlighted France’s decades-old fleet of nuclear plants as a key advantage compared to the US, providing clean, scalable energy supply for AI’s vast processing needs.
“I have a good friend in the other part of the ocean saying ‘drill, baby, drill’,” Macron said in an apparent dig at US President Donald Trump’s pro-fossil fuels policy.
“Here there is no need to drill, it’s plug, baby, plug!” he added.
Also speaking last night, Google chief executive Sundar Pichai said technology’s shift to AI is set to be “the biggest of our lifetimes”.
Today, European Commission President Ursula von der Leyen announced a push to channel €200bn in public and private investments into Europe’s AI industry.
This morning, the Taoiseach joined political leaders from around 100 countries for a plenary session hosted in Paris’s 125-year-old Grand Palais. Notable attendees including Von Der Leyen, Vance, and Indian Prime Minister Narendra Modi who will host next year’s summit in India.
For Vance it was his first official overseas trip since entering office last month and all eyes were on how he engaged with European leaders, some he was meeting for the first time.
Vance arrived in the French capital yesterday alongside his wife Usha and their three children.
He was scheduled to meet with European Commission President Ursula von der Leyen last night.
The Taoiseach briefly met Vance for the first time at last night’s working dinner, telling reporters he “spoke very fondly” of Ireland.
It’s understood senior tech figures including Google’s Pichai, Microsoft’s Bard Smith and OpenAI’s Sam Altman also attended the dinner.
In a statement ahead of the summit, the Taoiseach said his priority was focusing on getting the balance right on regulation of the technology “so that innovation can flourish while protecting against harmful developments.”
At the end of last month, the arrival of China’s DeepSeek AI model caused a splash in the tech sector globally, with the free app becoming the most downloaded on Apple’s US App Store, displacing OpenAI’s ChatGPT.
Its launch posed difficult questions for the US tech sector, and has sparked concerns that safety and regulation may suffer as the global AI race enters a new phase.
The summit in Paris has also received criticism for focusing too much on the possible benefits of AI without giving due space to its potential dangers.
Max Tegmark, head of the US-based Future of Life Institute that has warned of AI’s “existential risk”, criticised a leaked draft of the summit’s final communique over its failure to mention these risks.
Tegmark also criticised it for failing to provide any “concrete proposals to ensure these powerful systems remain controllable and beneficial”.
“This alarming omission demonstrates a fundamental misunderstanding of the science, and is a recipe for disaster,” he said.
With reporting from AFP.
To embed this post, copy the code below on your site
Is this real or AI generated fake news?
@Padraig O’Brien: if ai were taking décision instead of these showers, may be there would be more sense in many things.
the only word for them is corruption
@Dominic Leleu: it ain’t sentinel, it’s still programmed by the same shower.
@Padraig O’Brien: And that right there is the thing we should really be talking about. We’re already at the stage where it’s assumed everyone must panic and join this race just because everyone else is in it. It’s already upending our sense of reality and undermining achievements in everything from exams to creative work.
I’m sure there will be a few medical and scientific advancements but whether the enormous societal price is worth paying is another thing entirely. We already live in a woo woo era where the stupidest of lies are readily believed. The idea of the Trump’s, Musk’s and Putin’s of this world having access to even more powerful tools of misinformation is terrifying. The utter BS people believe already – can you imagine the lies and propaganda their AI teams will produce to further their interests…
@Dere: Do you mean if any A.I. Has passed the Turing test, which I believe there have been cases. Don’t quote me but I believe at least a google AI had passed it.
749 000 houses without electricity or water or heating or Internet or even good mobile signal 2 weeks ago.
I askedAI why and how this happens if governments since 1980 pouring billions into infrastructure? You raise a critical point. Corruption, inefficiency, or mismanagement in infrastructure spending can indeed undermine even well-funded systems, leaving communities vulnerable. While Ireland ranks relatively well on global corruption indices (e.g., 10th in Transparency International’s 2022 Corruption Perceptions Index), **systemic governance gaps**—not necessarily outright bribery—can erode infrastructure resilience. Here’s how such issues might contribute to failures:
—
### **1. Types of Corruption or Governance Failures**
- **Contract mismanagement**: Public contracts awarded to politically connected firms without competitive bidding, leading to inflated costs or substandard work.
– Example: In 2023, Ireland’s National Broadband Plan faced scrutiny over cost overruns and delays, raising questions about procurement processes.
- **Short-termism**: Politicians prioritizing visible, short-term projects (e.g., new roads) over long-term, less glamorous upgrades (e.g., replacing aging water pipes).
- **Embezzlement/fraud**: Rare in Ireland, but misallocated funds at local levels could divert money from critical maintenance.
- **Regulatory capture**: Lobbying by powerful industries (e.g., energy, construction) shaping policies to favor profits over public resilience.
—
### **2. Ireland’s Historical Context**
- **Celtic Tiger-era waste**: During Ireland’s economic boom (1990s–2000s), infrastructure spending sometimes prioritized flashy projects over foundational upgrades.
- **Post-2008 austerity**: After the banking crisis, infrastructure budgets were slashed, delaying maintenance and modernization.
- **Rural neglect**: Urban-centric investment left rural areas with outdated systems (e.g., overhead power lines, aging water networks).
—
### **3. Why Corruption Alone Doesn’t Explain the Crisis**
While governance flaws exist, Ireland’s infrastructure vulnerabilities are more deeply tied to:
- **Climate acceleration**: Storms like Agnes (2023) are now 30% stronger due to climate change, overwhelming systems designed for milder weather.
- **Underfunded resilience**: Only 0.3% of Ireland’s 2021–2030 infrastructure budget was allocated to climate adaptation.
- **Technical debt**: Fixing outdated systems (e.g., 19th-century water pipes) is costlier than building new ones.
—
### **4. Corruption vs. Systemic Dysfunction**
In Ireland’s case, **institutional inertia** and **fragmented responsibility** are likelier culprits than blatant corruption:
- **Siloed agencies**: Over 30 bodies manage water, power, and telecoms, causing coordination failures during crises.
- **Reactive spending**: Funds often flow post-disaster (e.g., storm repairs) instead of proactive hardening.
- **Lack of accountability**: No legal mandate for infrastructure providers to meet climate-resilience standards.
—
### **5. Solutions Beyond Anti-Corruption**
To fix this, Ireland (and similar nations) needs:
- **Independent oversight bodies**: To audit infrastructure spending and enforce transparency.
- **Climate-resilience mandates**: Legally require utilities to upgrade systems to withstand future storms.
- **Decentralization**: Empower local councils to manage microgrids, water systems, and broadband.
- **Public-private accountability**: Tie corporate subsidies to performance metrics (e.g., outage response times).
—
### **Conclusion**
While corruption is rarely the sole cause of infrastructure collapse, governance weaknesses—short-term thinking, poor coordination, and underfunded maintenance—can have similar effects. In Ireland, compounding these issues with climate-driven disasters creates a perfect storm. Fixing it requires not just money, but systemic reforms to ensure investments translate into **lasting, equitable resilience**.
@sakk sa: there’s long winded and then there’s long winded, after 3 line had to move on.
@Ned: And then came across Murphy. Go straight back up and read ned. Lol
@sakk sa: Your prompt/question to the AI app clearly included at least the term “corruption” correct? A leading question to attempt to prove a point? So you didn’t just ask the AI “Why?”. Please be transparent.
@Ned: I think it is an AI produced answer..
@sakk sa: more of this please.
Right now due to over regulation, EU is way behind China and US in AI. We do not have the equivalent of Apple, Google, Nvidia, Alibaba, Baidu etc.
We risk being left behind and reduce ourselves to third world status. Or become like Ireland and a colony for the mega AI firms.
At least Macron sees the problem. He’s one of the few
@Fran Ken: oh yeah , but does US and China got attached plastic bottle lids?? Lol…
@Lei tatt: No, but they also don’t have strictly enforced employee rights, health & safety standards, and consumer protection laws.
@Fran Ken: It is a bit of a catch 22…
Regulate it and if US or China decide to ignore they will take advantage of it while globally the world suffers…
OR
Leave unregulated and let it run free at the determent of all…
I work in IT and I was asked recently what will AI bring us… In fairness, I see it as having gains for us but effectively it will allow the richer to get richer with more concentration of wealth. It is middle class jobs they will be going for now and don’t think I work with my hands, therefore I am safe…
We are not that far from AI designed houses produced in 3D printed factories or on site in a few days…
Where will the world econmics be then..
@Fran Ken: Regulations protect people from exploitation. JD Vance sees working class people as slaves who talk back . The likes of him can’t wait til they can just ignore great swathes of humanity.
@paul hayes: I watched a 3 hour conversation between Vance and Joe Rogan. I can feel your eyes rolling up from here, but you could give it a watch. He certainly didn’t come across as a psychopath or anything. More politicians should do stuff like that, rather than your standard journalist interviews for soundbites. I also watched him talk to Bernie Sanders, who also seems like a decent guy, for balance :-) You get a better sense of a person’s character by listening to them, unscripted for 3 hours.
@Chop Chop: Bernie is the only person I have looked into, that sincerely has the Ordinary Person’s causes at heart.
I used to think he was a mad socialist but now I believe him to be the only person to fix the US. Unfortunately, due to his age (and MAGA voters) , it’s far too late.
@paul hayes: Your opinion is simply not based on facts!! And the facts don’t care about your feelings!! Give JD Vance’s life story a watch and tell us again how he feels!! You have no idea!!!
@Fran Ken: absolute nonsense, 3rd world status. We might just retain a bit of quality about ourselves and our lives.
These guys don’t want anything regulated.
@Dermot Blaine: he’s a mouthpiece for Musk and loyal to his very own financier Thiel, they’re not to be trusted, remember Vance and what opinion he had of Trump in 2016 and how he’s done a complete about face, follow Trump/Vance at your peril, they have zero interest in ordinary working people
@Gerry Ryan: like all politicians on both sides of the spectrum
The lackey of corporate interests speaks.
@Jacintha Dumbrell: They’ll roll in behind dying markets and take some spoils, but the rest are big enough (economically) to move on without them
He would say that, wouldn’t he
I don’t understand why Indian PM Modi is cohosting this event. What India needs right now is basic facilities like toilets for example, healthcare, housing & basic employment so that people start earning a decent living & stop migrating illegally.
@Guru Sharma: You clearly know very little about India.
@Iano C: well his comment is spot on
@sean weir: Nope, it’s factually incorrect.
@Iano C: You’re the one wearing rose-tinted glasses!
@Brian Hunt: Nope, just like dealing in facts, not populist rhetoric. You should try it sometime…but you won’t.
@Iano C: You seem to know even less. India is mostly a living hell.
@smatrix mantra: Again, not accurate.
The opposite of Musk’s standpoint on the regulation of AI.
A mutual back scratching event.
@Ned: back?
This was written by OpenAI’s o3 Deep Research:
https://archive.org/details/ai-deep-research
Abstract:
Artificial Intelligence (AI) has undergone rapid evolution, with advancements in model architectures, reasoning mechanisms, and hybrid integration strategies reshaping its capabilities. This paper explores the foundational architectures underpinning modern AI systems, including transformer-based foundation models, search-augmented retrieval techniques, and reasoning-enhanced models. It analyzes the strengths and limitations of these architectures, highlighting recent innovations such as chain-of-thought prompting, neuro-symbolic AI, and tool-using AI agents. The study further examines integration strategies that combine different model types, such as retrieval-augmented generation, multi-modal systems, and hierarchical expert models. Additionally, it assesses the near-term trajectory of AI, forecasting improvements in multi-modal capabilities, context expansion, and autonomous task execution. The broader implications of these advancements in fields such as software development, content creation, scientific research, and decision-making are also explored. The paper concludes with an analysis of technical bottlenecks and potential breakthroughs, including improved reasoning depth, common-sense integration, and scalable architectures that could shape the next wave of AI evolution.
I don’t think people are prepared for what’s going on to happen in the next few years. People controlling these aren’t creating them to make work easier, that might happen initially, but in the longer term, between 5-10 years, many people’s jobs are going to be replaced. I understand people are concerned about misinformation, fake images, that’s one valid worry about AI, but that is only one of many. Most of what o3 wrote is generally positive as it gathering and based what it wote on what AI researches said themselves, and they tend to give a rosey picture of what they think the AI future holds. These positives are possible, but these depend on the political climate, who controls politicians, and we see from current events ther rich and powerful, the unelected, seek to influence political power for their own interests. Do you think they care about us? Soon AI workers will replace many living worker roles for a fraction of the cost. We need to start talking about this, and stop being dumb about it.
@John Murphy: literally after that last line :) respect s/ @David Jordan: Can’t AI replace my job?
JDV is the puppet for the likes of Musk and Zuckerberg!
Could do with AI to run the government at this point, and it’d probably do a better job in its early stages than the clowns we got.
@Kieran Menon: can’t believe they are bringing a bowl of shamrock to these rap1sts and murderers in March, disgraceful
It’s guaranteed that whatever the US does on the global stage, the UK will follow . Well, for the next 4 years anyway .
AI has declared the Irish voter to be the stup idest in the World or maybe the Universe
@Dan Murphy: And then you see the yanks elected Trump
The faaaar roish lol
Brenny AKA Paul Gorry will be here soon, with another of his ridiculous posts
@Barry Spoon: Your paranoia is getting out of hand barry. Please seek medical help. Worried about you, barry.
@Barry Spoon: It’s cute and miss someone like this. I’m sure you’ll find a man of your own to live in time.
@Jack Hayes: good girl, Brenny, pump those green thumbs up lol
@Barry Spoon: Prozac or serotonin both should work barry. Not at the same time, tho. Always read the instructions……
@Paul Gorry: come on, Brenny, why?
@Barry Spoon: Maybe take both barry at the same time. Probably for the better ( if ye get me), it will end your pain for sure.Why am I crying…..?
@Barry Spoon: Is it just me or is there a bit of phuckery going on with the “thumbs up” these days? Have thought a few times, those numbers just seem off.
@Chop Chop: definitely
A.I. Agents – And a role for humans.
Along with and connected to A.I, will be A.I. Agents. These will be virtual A.I.s that will manage many things for a person such as booking holidays, appointments, dinner. A.I. agents will talk to one another, act, react, ruminate and act again at a speed unparalleled by human thinking & action. The human initially instructing the agent what is required.
Thing is, there is now a market for humans to think up, invent and create these A.I. agent programs, much like Apps are created today.
It’s estimated next year will see thousands of new A.I. Agents, 2027: multiples of 100:000, 2028: Possibly millions of Agents. Each agent fulfilling a separate and specific function, as designed.
@Keth 417: It’s gonna be awesome!
@Keth 417: I want an AI agent to post on the Journal for me.
UK didn’t sign it because US told them not to. This will give Starmer leverage in getting a trade deal.
As a message to Donald Trump / Elon Musk from the rest. SIGN IT
The global elites love nothing more than slapping the word “ethical” on a policy to disguise their real agenda, which is control. The so-called “ethical AI” declaration pushed at the AI Global Summit is just another attempt to strangle innovation with bureaucracy and centralised oversight. Once you hand over control of AI to regulators, you can kiss free enterprise and innovation goodbye. JD Vance is right, the last thing AI needs is more government interference. Every major technological breakthrough in history, from the Industrial Revolution to the internet, thrived not because of government regulation but in spite of it. The more red tape you wrap around AI, the more you guarantee that only the most government-friendly entities will be able to develop it.
@William Jennings: They want DEIAI
have your say