Tom Allison

Lead Mentor for UX #GoogleLaunchpad, UX strategist and team leader

[Tom Allison @tallison]

Gold.

NYTimes: “You play for your fans,” [Vanilla Ice] said. “And I’ll go play for Putin, and I’ll play in Iran if you want. It don’t matter.”

The senior administration official said the White House appreciated that Vanilla Ice still agreed to participate in the event.

[The Tennessee Holler @TheTNHoller]

Here’s a thought — Maybe these people don’t deserve to be handsomely rewarded?

[Tom Allison @tallison]

I think if we could truly come to understand all the AI models as "useful cognitive instruments with bounded reliability" it would solve 99% of the problems they create.

[Tom Allison @tallison]

I hope OpenAI is carefully monitoring whether pre-IPO optimization pressures around cost, latency, and scalability are unintentionally degrading carefulness, grounding, and ambiguity resolution. I’ve been noticing some very odd response-behavior in the last week. Anyone else?

[Tom Allison @tallison]

I’ve been noticing some really odd behavior on ChatGPT‘s part this week, then this dropped: https://t.co/k1fsWpRvoD. I wonder what metrics that behavior is tied to? ;)

[Tom Allison @tallison]

Oh #BVG, just when you almost made BER airport fully functional… We are just in from Barcelona and there’s no elevator (broken), no down escalator (apparently doesn’t exist?) AND ALL THE ELECTRONIC SIGNS ARE WRONG. Do you want us to walk into town?!

[Tom Allison @tallison]

If I were #Sixt I would sue whatever entity is responsible for my new AI-enhanced customer service system. It doesn’t work and ends up being hostile to customers. I don’t imagine that’s what they contracted for. I am definitely less likely to use their service, going forward.

[Tom Allison @tallison]

I’ve used #Sixt for many years in Berlin. I have just now had the worst customer service experience with them ever. I never spoke to a human but I did try to chat with an apparently insane AI bot and I was put on endless hold more than once. Talk about degradation of service!

[Tom Allison @tallison]

Is this the only real political question: What does a fair distribution of wealth in any and all human social systems look like?

[Tom Allison @tallison]

In the end, getting something useful out of GenAI is frustratingly hit-and-miss regardless of how boss-level ones prompt mojo is.

[Tom Allison @tallison]

Is AI’s tremendous ability to SEEM RIGHT a greater deficit than all its strengths combined? https://t.co/zgdAMFCUJf

[Tom Allison @tallison]

Even the latest AI can’t keep track of the Platform-Spaghetti that plagues the viewer of contemporary television content.

[Tom Allison @tallison]

We witnessed something absolutely beautiful and profoundly human as the astronauts on Artemis 2 shared a moment of pain and memorial just as they were at the pinnacle of such a massive achievement. The heart aches and spills over with pride simultaneously.

[Tom Allison @tallison]

Why has ChatGPT started rudely interrupting my more interesting chats with a very awkward and destructive user-interaction about their length having exceeded invisible limits and then throwing away already generated content in the process?

[Tom Allison @tallison]

Thrashing to War Pigs on a Sunday morning. A reminder that what is going on right now is not new - though perhaps a bit more naked and obvious? Does Elon even realize what a pig he has become?

[Tom Allison @tallison]

Though I would also currently be open to "simulated relationship" - as in "simulated companion", "simulated girl/boyfriend" - which might seem less dismissive of the real emotion that can be evoked on the human side of the dyad.

[Tom Allison @tallison]

I am a very strong proponent of labeling human-ai relationships as "artificial relationships". I believe its an important ethical and user-protective label. Other more neutral or positive names for the phenomenon are more inaccurate and/or intentionally manipulative.

[Tom Allison @tallison]

This is getting faaaaar closer to the reality of AI:

[Dan Williams @danwilliamsphil]

Some of my high-level beliefs about AI:
1. AI is an extremely abnormal, sui generis technology. We're building thinking machines! There are strong reasons to think that lessons drawn from previous technologies (e.g., about the pace of diffusion) won't apply, and that AI will pose countless novel challenges for which there is simply no precedent.
2. AI is not a new species. Until five minutes ago, all our examples of intelligent systems were living things. So when people think about AI, they typically—and typically without realising it—mistakenly import expectations about intelligent living things (e.g., where drives towards self-preservation and competition reflect their distinctive Darwinian design process) into their understanding of how AI is likely to behave.
3. It's a mistake to treat the "AI as normal technology" view and the "AI as potentially autonomous superintelligent alien species" worldview as the only perspectives. It's more likely that AI is an extremely powerful, extremely abnormal technology that will nevertheless not behave like a new species.
4. The main risks from AI come from misuse, not model misalignment. We should be more cynical about humanity and less cynical about AI. We know with certainty that human beings are often selfish, competitive, power-seeking, deceitful, and hypocritical. Releasing an extremely powerful technology into a world marked by vast inequalities of political and economic power among human beings is where most of the real challenges will arise. Many people aren't thinking clearly about this because their model of "human nature" is shaped by the range of behaviours they encounter in modern liberal democratic capitalist states, where self-interested incentives partly constrain the most sociopathic behaviours people are capable of.

[Dan Williams @danwilliamsphil]

Some of my high-level beliefs about AI:
1. AI is an extremely abnormal, sui generis technology. We're building thinking machines! There are strong reasons to think that lessons drawn from previous technologies (e.g., about the pace of diffusion) won't apply, and that AI will pose countless novel challenges for which there is simply no precedent.
2. AI is not a new species. Until five minutes ago, all our examples of intelligent systems were living things. So when people think about AI, they typically—and typically without realising it—mistakenly import expectations about intelligent living things (e.g., where drives towards self-preservation and competition reflect their distinctive Darwinian design process) into their understanding of how AI is likely to behave.
3. It's a mistake to treat the "AI as normal technology" view and the "AI as potentially autonomous superintelligent alien species" worldview as the only perspectives. It's more likely that AI is an extremely powerful, extremely abnormal technology that will nevertheless not behave like a new species.
4. The main risks from AI come from misuse, not model misalignment. We should be more cynical about humanity and less cynical about AI. We know with certainty that human beings are often selfish, competitive, power-seeking, deceitful, and hypocritical. Releasing an extremely powerful technology into a world marked by vast inequalities of political and economic power among human beings is where most of the real challenges will arise. Many people aren't thinking clearly about this because their model of "human nature" is shaped by the range of behaviours they encounter in modern liberal democratic capitalist states, where self-interested incentives partly constrain the most sociopathic behaviours people are capable of.