[ExI] Interesting take on AI.

[BillK]

On Sun, 30 Jun 2024 at 16:59, efc--- via extropy-chat
<extropy-chat at lists.extropy.org> wrote:
> Hello everyone,
> Thought you might enjoy this take on AI:
>
> https://techcrunch.com/2024/06/29/mit-robotics-pioneer-rodney-brooks-thinks-people-are-vastly-overestimating-generative-ai/
>
> "Brooks adds that there’s this mistaken belief, mostly thanks to Moore’s
> law, that there will always be exponential growth when it comes to
> technology — the idea that if ChatGPT 4 is this good, imagine what ChatGPT
> 5, 6 and 7 will be like. He sees this flaw in that logic, that tech
> doesn’t always grow exponentially, in spite of Moore’s law".
>
> More in the article above.
>
> Best regards, Daniel
> _______________________________________________


As Brooks is criticising AI, it only seems fair to give one of the
latest AIs the right of reply.  :)
BillK

Here is the response from Google's Gemma 2 AI -

This TechCrunch article reporting on Rodney Brooks' comments about
Generative AI presents a compelling but ultimately incomplete picture.
While it effectively highlights Brooks' skepticism, it lacks the depth
and nuance needed for a truly insightful analysis.

Strengths:

Attention-grabbing Headline: The title immediately captures attention
and sets the stage for a potentially controversial discussion.

Credibility: Leveraging Brooks' reputation as a leading robotics
expert lends weight to his criticisms and makes the article more
compelling.

Clear and Concise: The writing is straightforward and easy to
understand, making the topic accessible to a wider audience.

Weaknesses:

Oversimplification: The article tends to oversimplify Brooks'
argument, portraying him as simply dismissing Generative AI. While he
expresses skepticism, his critique is more nuanced, focusing on the
limitations of current AI and the dangers of overhyping its
capabilities.

Lack of Context: The article fails to adequately contextualize Brooks'
comments. Knowing the setting (a tech conference) and the intended
audience is crucial for understanding the tone and potential biases of
his remarks.

Limited Engagement with Brooks' Reasoning: The article touches upon
Brooks' concerns about the lack of true understanding and common sense
in current AI, but it doesn't delve deeply into the technical reasons
behind these limitations.

One-Sided Perspective: The article primarily focuses on Brooks'
criticisms without presenting counterarguments or perspectives from AI
researchers who might disagree with him. This lack of balance weakens
the article's analytical value.

Absence of Future Outlook: The article concludes without exploring
potential future developments in Generative AI. Does Brooks see any
paths for overcoming its current limitations? What advancements would
be needed?

Suggestions for Improvement:

Provide More Context: Explain the setting and audience of Brooks'
comments to help readers understand the context of his remarks.

Elaborate on Brooks' Reasoning: Dive deeper into the specific
technical challenges and limitations he sees in Generative AI.

Include Counterarguments: Present perspectives from AI researchers who
might disagree with Brooks, creating a more balanced and nuanced
discussion.

Explore Future Possibilities: Discuss potential paths for Generative
AI to overcome its limitations and what advancements might be needed.

Avoid Sensationalism: While the headline is attention-grabbing, the
article should avoid sensationalizing Brooks' views and instead strive
for a more objective and informative tone.

Overall:

The article serves as a good starting point for a conversation about
the limitations of Generative AI, but it needs further development to
provide a truly insightful and balanced analysis. By addressing its
weaknesses and incorporating the suggested improvements, the article
could become a more valuable contribution to the ongoing dialogue
surrounding AI's capabilities and potential.

----------------------------------