Parashat Yitro: I Am Not ‘Claude’
Before I begin, I would like to confirm that although I am Claude, I am not ‘Claude’, as per the subtitle of this Drasha. This does not mean that I have been replaced in an ‘Invasion of the Body Snatchers’ type operation. I have not lost control of my faculties. Far from it. It’s simple really. I’m just not ‘Claude’.
A short
while ago, a close friend introduced me to something that could potentially
change my life and perhaps yours. It
will certainly have an impact on the lives of your children and grandchildren
in ways that may not yet be fully understood. We are on the cusp of a transformative
technology which you might have heard of recently.
Friends,
I am of course referring to ‘ChatGPT’, the world’s most advanced AI, which we
know is an acronym for Artificial Intelligence.
The company behind it, OpenAi was formed in December 2015 by a number of
people including Elon Musk. Its founders
pledged $1 billion to develop its potential with Musk selling his share to
Microsoft three years later. Its
offerings first came to my attention with the release of DALL-E, a ‘deep
learning’ model that allows you to create digital images from natural language
inputs. So, for example, if I ask it to
draw ‘A penguin eating a banana whilst sitting in a swimming pool in the style
of Van Gogh, it will do exactly that.
This,
however, was only a precursor to the release of ChatGPT (which is an acronym
for Chat Generative Pre Trained Transformer) last November and was created
using a database containing the entire contents of the world wide web. It allows you to initiate a chat in the most
naturalistic manner available to date. In
other words, having used it, I can testify that it feels as though you could be
conversing with a human being. More than
that, you can ask it questions in the same way that you would to another person
and it responds, not only to your initial question but to the follow-ups. It is simply breath-taking in scope.
The only
way that I can adequately describe the power of ChatGPT is to share the result
that I received to the following question that I submitted into the chat box:
‘Write a sermon about Parashat Yitro comparing the
giving of the ten commandments to the use of an AI’
This is
the response that was instantly generated:
Dear friends,
in this week's Parshat Yitro, we read about the giving of the Ten Commandments
to the Israelites by God at Mount Sinai.
Just as the Israelites received these commandments as a guide for how to
live a just and moral life, we too must consider the guidelines we need to
create an ethical and responsible use of AI.
Just as the Ten Commandments
were meant to be followed in all circumstances and situations, the use of AI
must also be guided by clear ethical principles. These principles should encompass issues such
as privacy, transparency, accountability, and the responsible use of data.
Like the Ten Commandments,
these ethical principles for AI must be widely known and understood by those
who design and develop AI systems, as well as by those who use them. Just as the Israelites were called upon to
internalize and follow the commandments, we too must strive to embody these
ethical principles in all of our interactions with AI.
Moreover, the giving of the
Ten Commandments marked a momentous event in the history of the Israelites,
transforming them from a loosely organized group of individuals into a unified
community with a shared sense of purpose.
In a similar way, the adoption of ethical principles for AI can help to
bring together a diverse group of stakeholders, fostering a shared
understanding of the responsibilities associated with the development and use
of AI.
Finally, just as the Ten
Commandments have stood the test of time, enduring for thousands of years as a
timeless source of guidance and wisdom, so too must our ethical principles for
AI be able to endure over time and adapt to the changing circumstances of our
world.
In conclusion, let us remember
the lessons of Parashat Yitro as we consider the use of AI in our world. Just as the Israelites received the Ten
Commandments as a guide for how to live a just and moral life, so too must we
establish ethical principles for AI that can serve as a guide for how to create
and use this technology in a responsible and equitable way. May we always strive to use AI in a manner
that honors the dignity of all people and protects the rights and freedoms of
all. Amen.
When I
rephrased the question and asked it to include commentaries by Rashi and
Rambam, the response was even more astounding.
I have chosen not to include it as I wouldn’t want you to think that
this Drasha has been nearly entirely composed by the software!
As I am
sure you will appreciate, beyond the instant jaw-dropping response that you have
probably experienced, there is understandable alarm emanating from the academic
and literary worlds regarding the veracity of the printed word. In an age where great efforts have been made
to limit plagiarism, this tool is a godsend to anyone who is tasked with
writing an essay, thesis or book. How
would we know that it hasn’t been created by a robot? Amazon is currently selling a book called ‘The
Inner Life of an AI: A Memoir by ChatGPT’, released in the second week of
December– if you don’t believe me, you can check it out by clicking
Which is
where ‘Claude’ comes in.
Former
employees of OpenAi (or as we would call them ‘the breakaway minyan’) have
created a rival to ChatGPT called….you guessed it….’Claude’!
Anthropic’s
‘Claude’ is different to his virtual cousin in that he is built to be more
user-friendly and able to refuse certain inappropriate questions on ethical
grounds (I’ll let you imagine what these could be). He apparently has a better sense of humour in
his responses and provided very long and verbose answers to questions (so in
that way, he’s quite like me!).
You can
read a fascinating comparison of both AIs at
https://scale.com/blog/chatgpt-vs-claude#Text%20summarization
and
waiting in the wings is Google’s version which is called ‘Bard’.
You may
be wondering how all of this information ties in with this week’s Parasha
(beyond the answer provided by ChatGPT) and if I told you that I couldn’t think
of an adequate response, you might suggest that I simply refer to ‘Claude’s’
rival’s answer. However, that is
precisely the reason why I chose this week’s sedra to discuss this technology.
This Claude (i.e. me – are you confused yet?) does not need programming to recognise the authenticity of the Torah as the Divine word of Gd. The Lord didn’t create a breakaway minyan to talk to a few select Israelites in order to announce His presence to the world. We are the only nation on earth who received the Divine word en masse. Millions of our ancestors heard Hashem (mostly through the voice of Moshe) declare the ten (well, thirteen, but let’s leave that to one side for now) pronouncements that would become the bedrock of the world we live in.
Commandments
such as:
·
Don’t serve idols
·
Respect your parents
·
Recognise Gd’s labour and efforts in creating the
world by ensuring that we set aside a day of rest
·
Don’t steal or murder or destroy the holy union of
marriage through adultery
·
Don’t pervert the course of justice through lying
under oath
communicated
the message that, for society to exist and thrive, two relationships had to be
established and nurtured, that of human-to-human and human-to-Gd.
When we
respect each other, there is nothing we cannot achieve together. When people truly venerate Gd, irrespective
of the different ways they choose to do so, we also thrive. We have seen what happens when either of
these relationships fail.
As I
write these words, tragically, tens of thousands of people have been killed in
devastating earthquakes in Turkey and Syria.
Putting aside their differences and decades of hostility, Syria has
asked for help from Israel and Israel has not hesitated to offer her hand in
friendship. This is not something that
‘Claude’ or ‘ChatGPT’ could predict, even if they have the entire web flowing
through their machine code. This is what
it means to be a human, blessed with a soul and a heart.
With all
the conflicting emotions that make us who we are. All
the contradictions that defy logic and reason.
All the strengths and weaknesses
that define us and puzzle those who try to understand our thought processes.
This is
what makes me Claude and not ‘Claude’.
Artificial
Intelligence is intelligence but it is also artificial. By its very nature, it cannot be real, in the
same way that Virtual Reality can only ever be virtual.
Reading
through the Aseret Hadibrot (‘The Ten Pronouncements’) is a good place to
start.
Shavuah Tov.
Comments
Post a Comment