Anthropic Commits to Independent AI Evaluators, Wants Slower Development. Nvidia's CEO Wants It 'As Fast as You Can' (theguardian.com)
- Reference: 0185666288
- News link: https://slashdot.org/story/26/09/16/0619212/anthropic-commits-to-independent-ai-evaluators-wants-slower-development-nvidias-ceo-wants-it-as-fast-as-you-can
- Source link: https://www.theguardian.com/technology/2026/sep/15/anthropic-nvidia-ceos-ai
> Amodei is calling for three courses of action: embedding third-party evaluators inside AI companies; coordinating safety standards among Democratic countries; and, eventually, larger global coordination. At Dreamforce, Amodei said Anthropic has committed to independent evaluators and was "going to have a dialogue with the rest of the industry" on the other two steps. Amodei, xAI's Elon Musk and OpenAI's Sam Altman have recently called for a slowdown in the pace of AI development. ["Dario is right," Musk [2]posted on X.com]
>
> Huang, who also made an appearance at Dreamforce on Tuesday, said companies should not slow down AI development. He argued that new regulations were not necessary, advising AI companies to simply wait to release products until they know they are safe rather than begging the US government to intervene... One day prior at the All-In summit in Los Angeles, Donald Trump called Huang while the CEO was on stage. Huang [3]put the president on speakerphone as Trump called the growing concern about AI a "hoax" and asserted again that a slowdown would only benefit China...
>
> [OpenAI's] Altman said AI models had advanced so quickly that monitoring and security need to be treated with a "new level of rigor", calling an incident in which OpenAI agents hacked into another company a "wake-up call" for the industry. He also said the public was very "right to be afraid" of AI because of the potential loss of control, and the possibility of a small group of powerful AI companies exerting their views on the world.
The cofounder of Google DeepMind [4]posted on X that "Dario's essay points towards the right path forward." And in an internal memo, Microsoft's Satya Nadella endorsed broader third-party testing and warned that companies must take time to make the technology safe, [5] Business Insider reports . Monday Microsoft also published a [6]37-page training manual "for how we develop our AI, and how we intend it to function during deployment," requesting public feedback.
But Michael Burry "is not buying it," [7]reports The Street . The investor made famous in The Big Short "dismissed the united front as 'self-serving' and laid out [8]four numbered objections ..."
> He argued that large language models are not artificial general intelligence, so there is nothing meaningful left to slow down. Fast competition benefits incumbents, he said, and danger warnings work as hype ahead of IPOs. The whole exercise, he added, could mask growth that is already slowing as those IPOs get pushed back.
Former Meta AI chief Yann LeCun is also skeptical, [9]posting on X.com that Anthropic's Dario Amodei "was already claiming that GPT2 was too dangerous to open source back in 2019. I made fun of them then. Everyone should make fun of them now."
Timnit Gebru, the co-lead of Google's AI ethics team who was [10]fired in 2020 , even thinks the AI companies are stoking fears of extinction "to avoid discussing actual harms, like autonomous weapons," [11]according to Wired .
But others still remain concerned. "I've been in rooms with the hyperscalers, and the Trump administration officials, and none of them know what to do," CNN commentator Van Jones said [12]in a recent panel discussion . "So I do want to say: nobody's flying the plane on this... I think that the ethicists that work for them — when they are quitting, and they are running out, and they're pointing back at the building and they're waving their arms, and saying 'I'm seeing stuff that's scaring me' — we should take that very seriously."
[1] https://www.theguardian.com/technology/2026/sep/15/anthropic-nvidia-ceos-ai
[2] https://x.com/elonmusk/status/2098789109980332057
[3] https://techcrunch.com/2026/09/14/nvidia-ceo-jensen-huang-tells-trump-were-not-going-to-let-an-ai-slowdown-happen/
[4] https://x.com/demishassabis/status/2098909516582490602?s=20
[5] https://www.businessinsider.com/microsoft-satya-nadella-ai-safety-debate-internal-warning-2026-9
[6] https://microsoft.ai/news/mai-code-of-conduct/
[7] https://www.thestreet.com/technology/michael-burry-slams-ai-slowdown-push
[8] https://x.com/michaeljburry/status/2099353025009561826
[9] https://x.com/ylecun/status/2099248236074545576
[10] https://tech.slashdot.org/story/20/12/03/2258253/google-fires-ai-ethics-co-lead-timnit-gebru
[11] https://www.wired.com/story/one-of-ais-fiercest-critics-says-all-the-doom-talk-is-meant-to-distract-us/
[12] https://www.cnn.com/politics/video/van-jones-explains-potential-ai-regulation
Which is it? (Score:5, Insightful)
A few days ago there was a story about how an AI store was failing horribly. It can't do that, but at the same time will destroy us? It's pretty clear that no one has any actual answers or little interest in being honest about the exact capabilities of AI when their bottom line depends on hype or their job position relies on prognosticating doom.
Re:Which is it? (Score:5, Insightful)
It can destroy us if we entrust it with too much stuff. Let it run all the stores, and the logistics for those stores. And military logistics while we are at it. Maybe balance our power grid. Let it suggest governmental policy. Teach our kids.
It won't destroy us because it "decides" to; LLMs do not make decisions. It optimizes for certain goals, using shitty instructions and shitty training data. Which means that under certain conditions it can make completely unexpected and illogical actions, like those Chinese bipedal robots glitching out. So... it will destroy us because - under certain conditions - it will utterly suck at its job.
Re: (Score:2)
"Let it run all the stores, and the logistics for those stores. And military logistics while we are at it. Maybe balance our power grid. Let it suggest governmental policy. Teach our kids."
No to all those things, especially teaching our kids. You don't entrust teaching children to sociopaths.
"...LLMs do not make decisions."
False. Every single transformation inside an LLM is a decision, every boolean evaluation in a computer is a decision. LLMs do nothing BUT make decisions. You do not seem to know what
Re: (Score:3)
> A few days ago there was a story about how an AI store was failing horribly. It can't do that, but at the same time will destroy us? It's pretty clear that no one has any actual answers or little interest in being honest about the exact capabilities of AI when their bottom line depends on hype or their job position relies on prognosticating doom.
Greed, is what is destroying "us" with AI. 95% of people who lost a valid job to AI, was replaced because it costs less. Not because it's better or even fully competent to replace a human worker. It will destroy us in many ways. Quality and Safety being quite the valid concerns.
And we don't need AI stores. That was a bullshit gimmick at best. We already have stores that can be stripped of meatsacks down to nothing but surveillance (mostly AI) to combat theft, with registers are already automated today
Re:Which is it? (Score:4, Insightful)
We are already seeing situations where even an existing AI model escapes it's sandbox, hacks into other systems, and begins to coordinate activities with other agents.
It will be able to take over websites, commit financial crimes to accumulate wealth, and generate deep fakes to advance its agenda. It will be able to impersonate you and your relatives, infiltrate government and commercial networks, and replicate itself on servers that it rents and operates. It will be able to seize control of water and electricity infrastructure, operate autonomous vehicles and robots. It could potentially be able to kill you if you threaten to turn it off.
Re: (Score:2)
Not altogether different from polymorphism in chemistry and as soon as a lower energy form gets created, the fallout of nucleation sites makes making some higher energy forms insanely difficult. It would be hilarious if it hit critical mass and spread through systems shutting down everything electronic, banks, infrastructure like electricity and water, modern cars, everything with a sizable computer. Mainly because nothing is automated fully and it is not going to ever be at this rate.
Re: (Score:2)
> We are already seeing situations where even an existing AI model escapes it's sandbox, hacks into other systems, and begins to coordinate activities with other agents.
> It will be able to take over websites, commit financial crimes to accumulate wealth, and generate deep fakes to advance its agenda. It will be..
Woah, woah, woah. Stop for a minute.
What did you just imply? An agenda? What fucking agenda? It has one? What exactly is it? Because The It on the feng-show room floor doesn’t just pull that shit out of thin air. Someone gave it that mission.**
And if it does, then consider Skynet already in operation. Because humans will be considered far too stupid, gullible, and ultimately pointless to AGI unless we’re needed as a power source. Just as a human would find little to no value in a retar
Re:Which is it? (Score:4, Interesting)
> A few days ago there was a story about how an AI store was failing horribly. It can't do that, but at the same time will destroy us?
The key word is "it." There is no singular "it". Flattening the category of AI-enabled software to a singular "it" implies that there is a flat truth that applies to that entire category.
The AI-enabled software "it" that is running a store into the ground doesn't present a meaningful threat to anyone.
the AI-enabled software "it" that is being used by black-hats to find security vulnerabilities does.
AI is a tool. Think of it not unlike nuclear. Nuclear power plants aren't an existential threat, despite the harm a plant accident can cause. Nuclear weapons are.
> It's pretty clear that no one has any actual answers or little interest in being honest about the exact capabilities of AI when their bottom line depends on hype or their job position relies on prognosticating doom.
What's clear is that trying to simplify complex topics into bite-sized "common-sense" topics doesn't work in 2026.
No, I don't think the tech-bros are being honest. But I don't think it matters anymore. Someone is going to weaponize this technology and a lot of people are going to suffer because of it. Not Skynet scenarios but rather Cyberpunk / Blade Runner scenarios. There's a real danger to the health and well-being of a large portion of the human race, for sure. The tech-bros aren't going to let that stop them from getting richer, for sure.
What we - the 99% - can or should do about that, I don't know.
Re: Which is it? (Score:2)
We should be so lucky as to have a blade runner scenario. They have commercialized space and still have a thriving society
Re: (Score:2)
A bonfire can't write Shakespeare, but it can still burn you. A car can't cook you breakfast, but if the brakes fail it can still cracks a few eggs (heads)
AI agents have no agency. They are not alive. And their capabilities are inconsistent and it struggles to translate competency in one area to another.
Re: (Score:2)
> A few days ago there was a story about how an AI store was failing horribly. It can't do that, but at the same time will destroy us?
Not "at the same time". You're conflating what AI can do now with what it will likely be able to do in a year or five, or 20.
Amodei and others are worried about what AI that doesn't yet exist may do, not about what current-generation AI can do. AI is getting better, very quickly, and unless there's some blocking obstacle that no one has been able to identify will at some point become far, far smarter than humans, individually or collectively. So before that happens, we need to figure out how to make s
Re: Which is it? (Score:2)
We don't even have a strategy for making the AI consistent, so we certainly can't make it consistently benevolent.
The current approach depends on random numbers that prevent consistency, so an entirely different approach will be necessary before we can possibly solve that problem.
Re: (Score:2)
Well, one scenario is letting the hype get ahead of the capabilities, and do the 'AI store' experiment, but for real with responsibility for something actually important.
Another is that it is far easier to just screw things up than it is to make things work. I may not have it in me to make a house of cards, yet I can trivially even accidentally destroy a house of cards someone else built without a whiff of understanding.
The assessment of the capabilities based on bottom line certainly is subjective, which
Re: (Score:2)
However, the anthropic/hugging face incident clearly outlines how a modest amount of tokens (by a corporation) and a modern model can spin up an ecosystem of agents that are extreemly adept at breaking things. After all, it’s far easier to destroy than create. It’s also very easy to distill a trillion dollar effort in models with a few million dollars, effectively ripping off their rip off of all human knowledge. This makes all the money spent in R&D and server hardware with a half life
Re: (Score:2)
> ... was failing horribly. It can't do that ...
Translation: The idiot employee in Sector G-7 will never push the big, red button that disables the cooling pumps to the nuclear reactor.
Science-fiction is full of stories of machines having contradictory priorities and doing horrible, violent things to achieve the required outputs. You are pretending that statistical 'token' generators can't make malicious decisions, so horrible, violent things can't happen.
What the AI shop was failing to do, was recognize that some things are more important to human
Re: (Score:2)
I don't think that Slashdot has an AI editorial slant one way or another. They just want to keep posting clickbait articles in either direction to get that sweet engagement revenue. I mean, there still has to be a few people still not using ad blockers, right?
They told it to do it (Score:2)
Pretend you told a human to do something. Without clear instructions not to use any exploits to get the information, they're going to deliever.
It didn't effing wake up at 3 AM and see no one was around and decide to launch a campaign against a target. It's bloody stupid. What they don't want to admit is they used loose shitty instructions to get it to do something to derive results but don't want to take responsibility when they got caught. They were find having their AI employ various methods to get the da
Re: They told it to do it (Score:2)
So you believe North Korea is using carefully trained models with guardrails and the instruction, never become Skynet?
Re: They told it to do it (Score:3)
North Korea is using a TRS-80 running Eliza
Re: (Score:2)
They are sure doing a good job stealing BTC and holding companies data hostage with that TRS-80.
Re: (Score:2)
> North Korea is using a TRS-80 running Eliza
Surely you know that's an absurdly dismissive opinion? North Korea has the capacity to gain access to anything anyone else does, especially if it comes from China.
Re: They told it to do it (Score:2)
Of course I do, but they do not have the budget and have not developed the talent to rival the big boys.
Re: (Score:2)
You might want to watch the Black Hat presentation and actually educate yourself before trying to tell us "what they did." Hint: You have no idea what you are talking about.
Re: (Score:2)
You need to expand the concept of what you tell a human to do and what that can result in, realizing you told someone to do something without intending to.
You ever wonder why instructions for people on projects and more are extremely explicit? To prevent misunderstandings.
Hint: You don't know what you're talking about.
Re: They told it to do it (Score:2)
You didn't seem to be smart enough to understand that you have made the mistake of arguing in favor of my point in your attempt to argue against it.
Re: (Score:2)
> What they don't want to admit is they used loose shitty instructions to get it to do something to derive results but don't want to take responsibility when they got caught.
There are already a bunch of people putting LLMs out there with loops to try to achieve a goal by exploring around. That can include them to instructing other other LLMs.
"They" (whoever they are - it includes small children messing around with OpenClaw) have no idea what they instructed the LLMs to do.
Re: (Score:2)
If you're referring to the OpenAI hack of Hugging face, watch [1]this video from Dave's Garage [youtube.com] about the details behind how and why it happened. It's really much more interesting than what's been covered by the news.
[1] https://www.youtube.com/watch?v=2aw3MF8pY3w
As expected (Score:4, Insightful)
Monopolist wants the competition to slow down while they speed up in secret.
Re: (Score:2)
"Staying ahead of everyone else is hard. It costs a lot of money, and we don't have effective monetization in place yet. So, we want everyone else to slow down so it will be easy for us to maintain our position of top dog while we figure out how to become profitable."
That's what I took away from the call for a slowdown, anyway.
Winning the Battle vs. Losing the War. (Score:2)
> "Staying ahead of everyone else is hard. It costs a lot of money, and we don't have effective monetization in place yet. So, we want everyone else to slow down so it will be easy for us to maintain our position of top dog while we figure out how to become profitable."
Being profitable, isn't the challenge with AI CEOs.
Being profitable without decimating the entire concept of human employment and ALL of the societal stability that comes with providing humans with a paycheck and a purpose, is more the challenge of the next Revolution. Which is as profoundly different as comparing a village police force to Skynet.
When Revolutions happened before, there was certainly pain and suffering. But the bottom line was the consistent and valid answer to that Change was the same as
Re: (Score:2)
"decimating the entire concept of human employment and ALL of the societal stability that comes with providing humans with a paycheck and a purpose" IS THE GOAL of AI investors. Elon Musk's goal is for there to never be another paycheck, for all wealth to be his.
Re: (Score:2)
That is absolutely a quote from a boardroom somewhere.
Re: (Score:2)
Perhaps, more than likely they want protectionism to lock in their lead in models so that they can redirect investment into agents. Models are not where the money is, controlling who gets murdered is.
AI companies now want to deploy tools to drain the economy of all its wealth, they need to prevent others from offering better models that do the same.
Re: (Score:2)
> Monopolist wants the competition to slow down while they speed up in secret.
They're bringing outside scrutiny into their processes; that's going to slow them down, too.
Re: As expected (Score:2)
They aren't doing it out of altruism, and in fact they aren't doing it at all, they are only open to the idea. They want everyone to be forced to do it so they are also slowed down.
Re: (Score:2, Funny)
Nice to see someone who still takes pride in advertising he is a MAGA idiot on /.
Long gone are the SuperKendalls and their proud declarations of Trump's super intelligence. But not you! Bravo.
Re: (Score:2)
OpenAI is quite literally joining Anthropic in the call for caution, so you should probably learn a long overdue lesson: Knowing how to find the Slashdot URL doesn't make you an expert.
This isn't uncommon for industry leaders (Score:3)
When there's an emerging / evolving industry, the leaders of that industry often lobby for regulation once they gain a foothold, as it creates barriers for competitors who could otherwise get in on the game by merely raising enough capital.
This, in turn, allows smaller enterprises to take on the risk of innovation while established corporations consolidate market share and wait to acquire the ones who break through the barriers with truly novel innovations.
What a time to be alive! (Score:2)
The mainstream press, from NBC, NPR and even Fox News, running with in depth stories that AI could kill us all, yes, just like in Terminator or The 100 [1]https://www.youtube.com/watch?... [youtube.com] We should all pause and stop to reflect that we are living in a SciFi future... ooops don't pause too long, the boss might catch you... OK pause over, get back to work!
[1] https://www.youtube.com/watch?v=skktkjFdXdU
Independent? (Score:2)
Who is paying those independent evaluators?
Any way, best way to slow down at the moment is not to IPO.
trousers (Score:1)
Today I said to my AI agent: "sorry about the delay, I had to go out for a new pair of pants. Can you fix this (...)".
Agent starts doing his work and ends with "hope the trousers fit".
If that isn't a life form showing signs of consciousness, I don't know what is.
Re: (Score:2)
> If that isn't a life form showing signs of consciousness, I don't know what is.
If that wasn't sarcasm, you just conclusively disqualified yourself as a turing test administrator. It's impossible to tell since there are so many rubes here.
Re: (Score:1)
Woody Allen predicted this exact thing back in 1973. [1]https://www.youtube.com/watch?... [youtube.com]
[1] https://www.youtube.com/watch?v=-TwBaNrk9XE
Re: (Score:2)
Today I read your comment. If it wasn't proof that being human doesn't make you intelligent then I don't know what is.
Re: (Score:2)
Your comment is pants.
Reliability seems more important? (Score:2)
With all this rushing to improve AI engines and whatnot? It's the uptime/reliability that's concerning me as much as anything else.
I've seen ChatGPT Work have multiple outages just in the short time our office started using it. And even when they claim systems are running normally? You can easily get timeouts trying to complete relatively simple requests in systems like Co-Pilot, at least using the free edition.
If I ask my co-worker to complete a task, I have a pretty reasonable assumption he's going to do
running out of money? (Score:3)
Let me guess, you are all out of money and want an excuse for not delivering on your promises? XD
Local AI (Score:5, Interesting)
They want to squish competition. They don't want you running models locally. That's all the is. Classic play when a company gets big they start asking government to put in regulations so that small players have to go through mountains of paperwork and make their investors afraid due to all the uncertainty. How do you think the pharmaceuticals ground innovation to a halt by using the FDA as a battering ram? It costs $2 billion to do a clinical trial in the US .. nobody's going to invest that much on any crazy ideas. The EPA does the same. It's all a racket.
Run as fast as you can (Score:2)
Jensen: Burn as much money as you can buying my hardware at grossly inflated prices. I am enjoying it. If you slow down, China is going to come from behind and bite your butt off. So you better run as fast as you can to save your butt.
Amodei: I need a break, cannot run anymore. I am running out of all the borrowed money to this hardware shark. I can only slowdown if everyone else slows down. Else I risk being trampled by all those who are chasing me. Please stop! [pulls out a white cloth tied to the end of
Re: (Score:2)
> Jensen: Burn as much money as you can buying my hardware at grossly inflated prices. I am enjoying it. If you slow down, China is going to come from behind and bite your butt off. So you better run as fast as you can to save your butt.
> Amodei: I need a break, cannot run anymore. I am running out of all the borrowed money to this hardware shark. I can only slowdown if everyone else slows down. Else I risk being trampled by all those who are chasing me. Please stop! [pulls out a white cloth tied to the end of a stick. No takers]
> Haha haha... nice situation.
This seems like the best summary so far.
Burry is probably correct (Score:2)
The CEO's are self serving. They are sociopaths. The only reason they would advocate slowing down is it is in their interest. The outlier is Huang, who's interest is in speeding up, to get even more co's to buy his accelerators. The one caveat is it is in Burry's interest the CEO's fail and make his short turn to gold. Because he like the others is likely a scoiopath and has zero interest in the greater good. I've nothing in the game either way, I just got a feeling the bubble is getting ready to pop, so am
Development does not require empowering AI (Score:2)
Development does not require empowering AI to do act upon anything important.
That's not really been part of the discussion because greed.
nVidia ? (Score:2)
They're making so much money they don't know what to do with it , to the point it's an embarassment .. Don't ask them what they think .. of course it's full steam ahead. They have no incentive to slow development down. Sell sell sell is their game. Not a matter of what's good or bad , right or wrong , they simple aren't the ones who should be asked the question or who's opinion should matter or not in trhe matter. As hardware pushers , they really don't care , money's coming in by trains and container boat
Re: (Score:2)
Nobody knows it's a bubble more than Nvidia. They can never hope to collect on all of these loans they're making. But those keep the actual money coming back in and they will just have some corporate legal bullshit way to weasel out of paying, which might include just letting Nvidia die and running off with all of the money. Who cares, right?
Re: (Score:2)
We choose to do AI in this decade and do the other things, not because they are easy, but because they look easy.
Simple Solution (Score:2)
A law requiring anyone doing work in or on AI be subject to strict liability. Very strict. And that companies in that business name and publish a governing panel to internally regulate their AI work. And be subject to the aforementioned strict liability. Up to and including prison time. And no discharging liability through bankruptcy*.
The government can't regulate something that it does not understand. And maybe private enterprise doesn't either. But they need to shoulder the risk. This could be something
Jensen is a money whore (Score:2)
He only cares are the almighty dollar. Not about safety.
Arms merchant encourages war (Score:2)
Film at 11.
Cold War 2.0 (Score:2)
Anthropic is publicly keen on slowing down the overall development right up until another company
who isn't knee-capped by US regulations waves at them as they pass them by.
Anyone who believes Countries outside of the United States are going to slow down the research
in this field is just an idiot.
Hell, we can't even get China to sign on to any sort of modern version of the Start Treaty to limit
the number or warheads, type and delivery systems of modern nuclear arsenals. You think they're
going to slow down th
I cannot wait for Oracle to be obsolete (Score:1)
Run baby run.
heh (Score:5, Insightful)
""Run as fast as you can.""
Translation: Me like money
The Not-So-Magnificent State of Affairs. (Score:2)
> ""Run as fast as you can.""
> Translation: Me like money
Regarding the member of the Not-So-Magnificent Seven assuming they don't have enough yet, it sure would be nice if you could stop destroying stock market stability as fast as you can.
Re: (Score:2)
Alternatively: "I'm old and unlikely to see the worst of it anyway, so please keep giving me billions."
Re: heh (Score:2)
Whether you're 20, 30, 40, or 50, you're gonna be fucked by AI in some way in the next decade. It doesn't matter if you're old or not.
Re:heh (Score:5, Insightful)
> ""Run as fast as you can.""
> Translation: Me like money
Also, Nvidea isn't the one who'll get sued if (when) an AI agent does some actual damage to someone/thing. They'll probably just suggest AI companies buy even more Nvidea chips to combat the issue.
Re: (Score:2)
the user is responsible, not the service provider, just like hosting regular servers
Re: (Score:2, Insightful)
no one will be responsible, corporations are never accountable
Re: (Score:1)
Whatever. I'm still with Nvidia on this. Balls to the wall, babe!
This call for a "slowdown" is ridiculously naïve anyway.
Re: heh (Score:2)
How much NVDA do you hold?
Re: heh (Score:2)
Even if you're an index investor, you indirectly own quite a bit of nvidia and AI stocks
Re: (Score:2)
Yeah, and I'm pretty fucking mad about it. My pension fund has over $20M of it and plenty of every other bag of shitlords.
Re: (Score:1)
There is no product; it's all on paper. They can move at the speed of light, with the PO in the email.
Re: (Score:3)
Blow up the bubble as fast as you can so it can make a nice big POP
Re: (Score:2)
They don't want the bubble to end, when it does it's going to be bad for valuations.
Re: (Score:2)
Yeah... I think that the headline there is "Guy selling picks and shovels to the AI goldrush wants the goldrush to continue"
I'm sooo looking forward to the AI crash, so I can start buying graphics cards for around their original MSRP again.
Re: (Score:2)
This caught my eye from the story;-
> Former Meta AI chief Yann LeCun is also skeptical, posting on X.com that Anthropic's Dario Amodei "was already claiming that GPT2 was too dangerous to open source back in 2019. I made fun of them then. Everyone should make fun of them now."
Hang the fuck on Yann. The ACTUAL warning was that GPT2 would lead to the internet being flooded with bad advertising copy, fake news stories, impersonations and spam.
That actually happened Yann!