Νέα

AI - Ανακαλύψεις & Προοπτικές

  • Μέλος που άνοιξε το νήμα Rand0m1
  • Ημερομηνία ανοίγματος
  • Απαντήσεις 12
  • Εμφανίσεις 622
  • Tagged users Καμία
  • Βλέπουν το thread αυτή τη στιγμή 1 άτομα (0 μέλη και 1 επισκέπτες)

Rand0m1

Σπουδαίος
Εγγρ.
15 Αυγ 2019
Μηνύματα
6.511
Κριτικές
1
Like
6.219
Πόντοι
2.356
Δεν ξέρω αν υπάρχει κάτι ήδη - δεν είναι κυριακάτικο πάντως! :confused:

Είπα ότι αξίζει ένα νέο thread, μια και συμβαίνουν πολλά τα τελευταία χρόνια.

Ας βάλω ένα βιντεάκι που είδα τις προάλλες - είναι εγγύηση ο Lex Fridman μια και έχει PhD στον τομέα.

 

Eraser..

Σπουδαίος
Εγγρ.
25 Δεκ 2016
Μηνύματα
28.790
Like
11.706
Πόντοι
2.606
εμείς πάντως την χρησιμοποιύμε ήδη στο νήμα των παππούληδων, όταν χρειαζόμαστε απαντήσεις σε δύσκολα θέματα

@blizardeye
 
Τελευταία τροποποίηση:

ΤΙΜΩΡΟΣ(1)

Ενεργό Μέλος
Εγγρ.
22 Οκτ 2009
Μηνύματα
1.280
Κριτικές
3
Like
3.084
Πόντοι
456
Δεν ξέρω αν υπάρχει κάτι ήδη - δεν είναι κυριακάτικο πάντως! :confused:

Είπα ότι αξίζει ένα νέο thread, μια και συμβαίνουν πολλά τα τελευταία χρόνια.

Ας βάλω ένα βιντεάκι που είδα τις προάλλες - είναι εγγύηση ο Lex Fridman μια και έχει PhD στον τομέα.

Ένα ρεζουμέ δε μας κάνεις ρε τύπε γιατί 5 ώρες βίντεο αποκλείεται να δω.
 
OP
OP
Rand0m1

Rand0m1

Σπουδαίος
Εγγρ.
15 Αυγ 2019
Μηνύματα
6.511
Κριτικές
1
Like
6.219
Πόντοι
2.356
Μοντέλα Τεχνητής Νοημοσύνης, τύπε.

Συγκεκριμένα, είναι μοντέλα φυσικής επεξεργασίας γλώσσας (NLP) τα οποία χρησιμοποιούν την αρχιτεκτονική των νευρωνικών δικτύων (NNs), για να δημιουργούν απαντήσεις σε ερωτήματα μέσω chatbots.

Βάλε εδώ ότι γουστάρεις και θα πάθεις την πλάκα σου!

 

ΤΙΜΩΡΟΣ(1)

Ενεργό Μέλος
Εγγρ.
22 Οκτ 2009
Μηνύματα
1.280
Κριτικές
3
Like
3.084
Πόντοι
456
Μοντέλα Τεχνητής Νοημοσύνης, τύπε.

Συγκεκριμένα, είναι μοντέλα φυσικής επεξεργασίας γλώσσας (NLP) τα οποία χρησιμοποιούν την αρχιτεκτονική των νευρωνικών δικτύων (NNs), για να δημιουργούν απαντήσεις σε ερωτήματα μέσω chatbots.

Βάλε εδώ ότι γουστάρεις και θα πάθεις την πλάκα σου!

Το ξέρω τύπε το chatgpt και το χρησιμοποιώ καθημερινά.
Το βίντεο ρωτάω τι λέει.
Γιατί έχω ακούσει πολλές μαλακίες, ότι θα μας πάρει τις δουλειές κλπ.
Είναι καλό εργαλείο αλλά δεν μπορεί να αντικαταστήσει τον άνθρωπο.
Γιατί πολύ απλά δεν έχει κριτική σκέψη. Αυτό που σου λέει, δεν ξέρει αν είναι σωστό ή όχι, και πολλές φορές λέει μπαρούφες, γιαυτό χρειάζεται ο άνθρωπος από πάνω.
Στην ουσία ένα autocomplete είναι αλλά πολύ χρήσιμο, δεν αντιλέγω.
Αλλά ούτε και στα καλλιτεχνικά μπορεί να αντικαταστήσει τον καλό καλλιτέχνη.

Επίσης και η σκοτεινή πλευρά είναι το πώς μπορεί να χρησιμοποιηθεί για παρακολούθηση και στον πόλεμο, εκεί πραγματικά είναι πάρα πολύ χρήσιμο. Στέλνεις 100 drones με AI και τα 10 να κάνουν μαλακία τα άλλα 90 θα πετύχουν όμως.
Επίσης στην παρακολούθηση όπως η microsoft που θέλει να επιβλέπει τι κάνεις αλλά και σε περιβάλλοντα εργασίας έχει προταθεί να παρακολουθεί όλες τις κινήσεις σου και να ενημερώνει το manager αν είσαι αργόσχολος. Μεγάλη μαλακία, θα κάθεται ο κόσμος να κουνάει τα ποντίκια και να γράφει μαλακίες για να φαίνεται παραγωγικός αντί να δουλεύει σωστά.
 
OP
OP
Rand0m1

Rand0m1

Σπουδαίος
Εγγρ.
15 Αυγ 2019
Μηνύματα
6.511
Κριτικές
1
Like
6.219
Πόντοι
2.356
Το ξέρω τύπε το chatgpt και το χρησιμοποιώ καθημερινά.
Το βίντεο ρωτάω τι λέει.
Γιατί έχω ακούσει πολλές μαλακίες, ότι θα μας πάρει τις δουλειές κλπ.
Είναι καλό εργαλείο αλλά δεν μπορεί να αντικαταστήσει τον άνθρωπο.
Έκανα την ίδια ερώτηση λίγο πολύ στο ChatGPT (δηλ αν θα μας πάρει τις δουλειές) και δες πως απαντάει. Δες και τη 2η ερώτηση που του κάνω.

Is AI going to kill all IT jobs?

ChatGPT said:
ChatGPT

AI is unlikely to "kill all IT jobs," but it will certainly transform the nature of many roles within the field. While automation and AI tools can handle repetitive, mundane tasks, there are several reasons why IT jobs will continue to exist, and in some cases, even grow:

1. AI Requires Human Oversight:
AI and automation tools need human supervision for development, maintenance, and ethical considerations. For instance, AI models need to be trained, fine-tuned, and monitored, which requires skilled IT professionals like data scientists, machine learning engineers, and AI specialists.

2. Emerging Fields:
As AI automates routine tasks, new fields in IT are emerging. Roles related to AI, cloud computing, cybersecurity, and data privacy are in high demand and will continue to grow. As companies adopt AI, they need professionals who can integrate, optimize, and secure AI systems.

3. Complexity of AI:
AI technologies are still far from being fully autonomous. They require deep understanding and ongoing adjustments. For example, IT professionals will still be needed for designing complex systems, ensuring their robustness, and adapting them to specific business needs.

4. Ethical and Legal Challenges:
As AI becomes more integrated into various industries, issues related to data privacy, algorithmic bias, and the ethical use of technology become more important. IT experts, especially in areas like AI ethics and law, will be in demand to address these concerns.

5. AI's Role in Augmenting IT Jobs:
Instead of replacing IT professionals, AI is more likely to act as a tool that augments their work. Tasks like system monitoring, threat detection, and troubleshooting can be enhanced by AI, allowing IT professionals to focus on more complex problems and strategic tasks.

6. Adaptation and Upskilling:
The IT field is accustomed to change. Historically, new technologies (like cloud computing or DevOps) have created job opportunities. As AI evolves, the key to staying relevant in IT will be continuous learning and adaptation. Professionals can upskill by learning about AI, automation, and advanced technologies to stay ahead of the curve.

Conclusion:
While AI will undoubtedly automate certain IT tasks and change how work is done, it will also create new opportunities, and the demand for skilled professionals who can design, implement, and manage AI systems will continue. Rather than eliminating IT jobs, AI will reshape the industry, requiring workers to adapt and evolve their skills.

You said:
Hopefully, IT developers will not be made redundant.

ChatGPT said:
ChatGPT

You're right to hope that IT developers won't be made redundant, and in fact, the outlook for developers is more promising than it might seem at first glance. While AI and automation are transforming the industry, they are unlikely to make developers obsolete. Instead, developers are likely to see their roles evolve in ways that can be both exciting and rewarding. Here's why:

1. AI as a Tool for Developers:
Rather than replacing developers, AI can enhance their work by automating repetitive tasks, like debugging, code generation, and testing. This can free up developers to focus on higher-level tasks such as problem-solving, architecture design, and creating innovative solutions. Tools like GitHub Copilot, for example, already assist developers in writing code faster and more efficiently, allowing them to work smarter, not harder.

2. Complex Problem-Solving:
AI may be able to write code, but it still struggles with understanding the context of a business problem or designing innovative solutions. Developers are crucial for identifying user needs, designing user-friendly interfaces, and creating solutions that align with business objectives. These creative and strategic tasks require human expertise.

3. AI Needs Human Supervision:
AI models and automation tools require continuous oversight and tuning to ensure they are functioning correctly. Developers will still play a central role in developing, refining, and maintaining AI systems, ensuring they work within specific organizational contexts and addressing issues like bias, security, and performance.

4. New Roles and Specializations:
As AI technology evolves, new job opportunities for developers will emerge. For instance, there will be a growing demand for professionals specializing in AI integration, AI ethics, machine learning, and data science. Developers with expertise in these areas will be crucial for helping businesses integrate AI technologies effectively and responsibly.

5. Upskilling Opportunities:
Developers who continue to upskill and stay updated on emerging technologies, including AI and machine learning, will be in a strong position. Learning how to build, train, and optimize AI models, or how to create tools that work alongside AI, will keep developers at the forefront of innovation.

6. Collaboration with AI:
Instead of replacing developers, AI will collaborate with them. For instance, AI can help developers write code more efficiently, but developers will still need to define the problem, choose the right algorithms, and apply creativity and domain-specific knowledge. This collaboration between human developers and AI tools will lead to faster and more sophisticated software development.

Conclusion:
Rather than rendering developers redundant, AI is set to reshape the role of IT professionals, including developers. The future will likely involve more collaboration between human developers and AI tools, making the development process faster, more efficient, and more innovative. Developers who adapt by learning new skills and embracing the tools that AI offers will remain at the core of the tech industry.

You said:
Thank you.

ChatGPT said:
ChatGPT

You're very welcome! I'm glad I could help. If you have any more questions in the future, feel free to reach out. Best of luck with your journey in the tech world!
-------------------------------------------------------------------------
Hello!

ChatGPT said:
ChatGPT

Hello! How can I assist you today?

You said:
I was wondering... has ChatGPT passed the Turing test yet?

ChatGPT said:
ChatGPT

The Turing Test, proposed by British mathematician and computer scientist Alan Turing in 1950, is a measure of a machine's ability to exhibit intelligent behavior indistinguishable from that of a human. The test involves a human evaluator interacting with both a machine and a human through text-based communication. If the evaluator cannot reliably distinguish between the machine and the human, the machine is considered to have passed the test.

As of now, ChatGPT, like other large language models, has made significant advancements in natural language processing and can often produce responses that seem human-like. However, whether ChatGPT has "passed" the Turing Test is somewhat subjective and context-dependent. Here are a few reasons why:

Context and Complexity: ChatGPT can carry on conversations and answer questions with considerable fluency on a wide range of topics. In certain situations, it might appear human-like enough to pass a Turing Test, especially in short or specific interactions. However, in longer or more complex conversations, the model can still show limitations, such as inconsistencies, a lack of deep understanding, or responses that feel robotic.

Evaluation Variability: The Turing Test is not a clear-cut, universally agreed-upon metric. Different evaluators might have different thresholds for determining whether a machine has passed the test. For example, in more casual conversations, ChatGPT might impress many users, but in others that require deeper reasoning or emotional nuance, it could still be identified as an AI.

Human-Like Conversations: While ChatGPT can generate human-like responses, it lacks genuine understanding, emotions, or consciousness. This means that although it may pass for a human in certain contexts, it doesn’t truly "understand" the conversation in the way a human does, which makes its performance fall short of true human-like intelligence in many cases.

In summary, ChatGPT can sometimes pass the Turing Test depending on the nature of the conversation, but it's not perfect and may still be identified as an AI by a discerning evaluator. So, while it demonstrates impressive conversational abilities, it doesn't fully meet the criteria for human-level intelligence as envisioned by Turing.

You said:
Sounds like a fair assessment. Thank you!

ChatGPT said:
ChatGPT

You're very welcome! I'm glad you found the explanation helpful. If you have any more questions or want to dive deeper into any topic, feel free to ask!
 
OP
OP
Rand0m1

Rand0m1

Σπουδαίος
Εγγρ.
15 Αυγ 2019
Μηνύματα
6.511
Κριτικές
1
Like
6.219
Πόντοι
2.356
Επίσης και η σκοτεινή πλευρά είναι το πώς μπορεί να χρησιμοποιηθεί για παρακολούθηση και στον πόλεμο, εκεί πραγματικά είναι πάρα πολύ χρήσιμο. Στέλνεις 100 drones με AI και τα 10 να κάνουν μαλακία τα άλλα 90 θα πετύχουν όμως.
Ως γνωστόν, η αμυντική βιομηχανία είναι αυτή που πάντα οδηγεί την κούρσα των ανακαλύψεων/επενδύσεων κ.ο.κ.

Δες 2 ωραία βιντεάκια που αναφέρονται και σε αυτό.


 

blizardeye

Τιμημένος
Εγγρ.
27 Δεκ 2013
Μηνύματα
81.718
Κριτικές
40
Like
25.411
Πόντοι
26.096

blizardeye

Τιμημένος
Εγγρ.
27 Δεκ 2013
Μηνύματα
81.718
Κριτικές
40
Like
25.411
Πόντοι
26.096
Έκανα την ίδια ερώτηση λίγο πολύ στο ChatGPT (δηλ αν θα μας πάρει τις δουλειές) και δες πως απαντάει. Δες και τη 2η ερώτηση που του κάνω.

Is AI going to kill all IT jobs?

ChatGPT said:
ChatGPT

AI is unlikely to "kill all IT jobs," but it will certainly transform the nature of many roles within the field. While automation and AI tools can handle repetitive, mundane tasks, there are several reasons why IT jobs will continue to exist, and in some cases, even grow:

1. AI Requires Human Oversight:
AI and automation tools need human supervision for development, maintenance, and ethical considerations. For instance, AI models need to be trained, fine-tuned, and monitored, which requires skilled IT professionals like data scientists, machine learning engineers, and AI specialists.

2. Emerging Fields:
As AI automates routine tasks, new fields in IT are emerging. Roles related to AI, cloud computing, cybersecurity, and data privacy are in high demand and will continue to grow. As companies adopt AI, they need professionals who can integrate, optimize, and secure AI systems.

3. Complexity of AI:
AI technologies are still far from being fully autonomous. They require deep understanding and ongoing adjustments. For example, IT professionals will still be needed for designing complex systems, ensuring their robustness, and adapting them to specific business needs.

4. Ethical and Legal Challenges:
As AI becomes more integrated into various industries, issues related to data privacy, algorithmic bias, and the ethical use of technology become more important. IT experts, especially in areas like AI ethics and law, will be in demand to address these concerns.

5. AI's Role in Augmenting IT Jobs:
Instead of replacing IT professionals, AI is more likely to act as a tool that augments their work. Tasks like system monitoring, threat detection, and troubleshooting can be enhanced by AI, allowing IT professionals to focus on more complex problems and strategic tasks.

6. Adaptation and Upskilling:
The IT field is accustomed to change. Historically, new technologies (like cloud computing or DevOps) have created job opportunities. As AI evolves, the key to staying relevant in IT will be continuous learning and adaptation. Professionals can upskill by learning about AI, automation, and advanced technologies to stay ahead of the curve.

Conclusion:
While AI will undoubtedly automate certain IT tasks and change how work is done, it will also create new opportunities, and the demand for skilled professionals who can design, implement, and manage AI systems will continue. Rather than eliminating IT jobs, AI will reshape the industry, requiring workers to adapt and evolve their skills.

You said:
Hopefully, IT developers will not be made redundant.

ChatGPT said:
ChatGPT

You're right to hope that IT developers won't be made redundant, and in fact, the outlook for developers is more promising than it might seem at first glance. While AI and automation are transforming the industry, they are unlikely to make developers obsolete. Instead, developers are likely to see their roles evolve in ways that can be both exciting and rewarding. Here's why:

1. AI as a Tool for Developers:
Rather than replacing developers, AI can enhance their work by automating repetitive tasks, like debugging, code generation, and testing. This can free up developers to focus on higher-level tasks such as problem-solving, architecture design, and creating innovative solutions. Tools like GitHub Copilot, for example, already assist developers in writing code faster and more efficiently, allowing them to work smarter, not harder.

2. Complex Problem-Solving:
AI may be able to write code, but it still struggles with understanding the context of a business problem or designing innovative solutions. Developers are crucial for identifying user needs, designing user-friendly interfaces, and creating solutions that align with business objectives. These creative and strategic tasks require human expertise.

3. AI Needs Human Supervision:
AI models and automation tools require continuous oversight and tuning to ensure they are functioning correctly. Developers will still play a central role in developing, refining, and maintaining AI systems, ensuring they work within specific organizational contexts and addressing issues like bias, security, and performance.

4. New Roles and Specializations:
As AI technology evolves, new job opportunities for developers will emerge. For instance, there will be a growing demand for professionals specializing in AI integration, AI ethics, machine learning, and data science. Developers with expertise in these areas will be crucial for helping businesses integrate AI technologies effectively and responsibly.

5. Upskilling Opportunities:
Developers who continue to upskill and stay updated on emerging technologies, including AI and machine learning, will be in a strong position. Learning how to build, train, and optimize AI models, or how to create tools that work alongside AI, will keep developers at the forefront of innovation.

6. Collaboration with AI:
Instead of replacing developers, AI will collaborate with them. For instance, AI can help developers write code more efficiently, but developers will still need to define the problem, choose the right algorithms, and apply creativity and domain-specific knowledge. This collaboration between human developers and AI tools will lead to faster and more sophisticated software development.

Conclusion:
Rather than rendering developers redundant, AI is set to reshape the role of IT professionals, including developers. The future will likely involve more collaboration between human developers and AI tools, making the development process faster, more efficient, and more innovative. Developers who adapt by learning new skills and embracing the tools that AI offers will remain at the core of the tech industry.

You said:
Thank you.

ChatGPT said:
ChatGPT

You're very welcome! I'm glad I could help. If you have any more questions in the future, feel free to reach out. Best of luck with your journey in the tech world!
-------------------------------------------------------------------------
Hello!

ChatGPT said:
ChatGPT

Hello! How can I assist you today?

You said:
I was wondering... has ChatGPT passed the Turing test yet?

ChatGPT said:
ChatGPT

The Turing Test, proposed by British mathematician and computer scientist Alan Turing in 1950, is a measure of a machine's ability to exhibit intelligent behavior indistinguishable from that of a human. The test involves a human evaluator interacting with both a machine and a human through text-based communication. If the evaluator cannot reliably distinguish between the machine and the human, the machine is considered to have passed the test.

As of now, ChatGPT, like other large language models, has made significant advancements in natural language processing and can often produce responses that seem human-like. However, whether ChatGPT has "passed" the Turing Test is somewhat subjective and context-dependent. Here are a few reasons why:

Context and Complexity: ChatGPT can carry on conversations and answer questions with considerable fluency on a wide range of topics. In certain situations, it might appear human-like enough to pass a Turing Test, especially in short or specific interactions. However, in longer or more complex conversations, the model can still show limitations, such as inconsistencies, a lack of deep understanding, or responses that feel robotic.

Evaluation Variability: The Turing Test is not a clear-cut, universally agreed-upon metric. Different evaluators might have different thresholds for determining whether a machine has passed the test. For example, in more casual conversations, ChatGPT might impress many users, but in others that require deeper reasoning or emotional nuance, it could still be identified as an AI.

Human-Like Conversations: While ChatGPT can generate human-like responses, it lacks genuine understanding, emotions, or consciousness. This means that although it may pass for a human in certain contexts, it doesn’t truly "understand" the conversation in the way a human does, which makes its performance fall short of true human-like intelligence in many cases.

In summary, ChatGPT can sometimes pass the Turing Test depending on the nature of the conversation, but it's not perfect and may still be identified as an AI by a discerning evaluator. So, while it demonstrates impressive conversational abilities, it doesn't fully meet the criteria for human-level intelligence as envisioned by Turing.

You said:
Sounds like a fair assessment. Thank you!

ChatGPT said:
ChatGPT

You're very welcome! I'm glad you found the explanation helpful. If you have any more questions or want to dive deeper into any topic, feel free to ask!
Ανέγνωσα προσεκτικά την ανάρτησή σας, κύριε.
Εξαιρετικά τα Αγγλικά σας εις τας ερωτήσεις!

Είστε καλοδεχούμενος στο Νήμα των παππούληδων!!!
 
OP
OP
Rand0m1

Rand0m1

Σπουδαίος
Εγγρ.
15 Αυγ 2019
Μηνύματα
6.511
Κριτικές
1
Like
6.219
Πόντοι
2.356
Άλλα 2 βιντεάκια... με τον Jensen Huang που μιλάει για σημαντικά πράγματα που έρχονται, όπως agenting AI models κλπ.


 

Ντουνταράς

Σεβαστός
Εγγρ.
4 Μαρ 2020
Μηνύματα
3.401
Κριτικές
11
Like
7.253
Πόντοι
1.346
Η Τεχνητή Νοημοσύνη μπορεί να αποτελέσει ένα χρήσιμο εργαλείο για όσους θέλουν να ασχοληθούν.
Αλλά η πραγματικότητα είναι ότι είμαστε τσοντάδικο κι όχι κανένα επιστημονικό forum.
Ο υποφαινόμενος δεν χρησιμοποιεί αυτή κάθε αυτή τεχνητή νοημοσύνη @dimigoall @MrEco με απώτερο στόχο να πάει στα κορίτσια περισσότερες φορές, αλλά έχει διαβάσει μερικές διπλωματικές στα ελληνικά όπως και κάποια ιστολόγια του εξωτερικού για να δει που περίπου κινούνται τα όρια επιτυχίας.

ΨΨ @Kompotis ως νεώτερος στρώσου κι άσε τα υπόλοιπα εδώ μέσα. Ο τοίχος είναι εκεί γύρω στο 70% για 1Χ2 ή πες 25-30% μεσοπρόθεσμο κέρδος στο κεφάλαιο που βάζεις.

ΨΨ @athinaios44 πάω στα Ρωσσίδια κανόνισε να με σπιουνέψεις ΓΚΡ...
 

Συνημμένα

  • Screenshot_20241209-082347.png
    Screenshot_20241209-082347.png
    144,1 KB · Εμφανίσεις: 8
  • Screenshot_20241209-082134.png
    Screenshot_20241209-082134.png
    158,6 KB · Εμφανίσεις: 10

Stories

Νέο!

Stories

Top Bottom