Try out our new FREE Youtube Summarizer!

Exploring Ethics and Safety

Harnessing NLP: Ethical and Secure AI Integration in Education and Workplace

Estimated read time: 1:20

    AI is evolving every day. Don't fall behind.

    Join 50,000+ readers learning how to use AI in just 5 minutes daily.

    Completely free, unsubscribe at any time.

    Summary

    The transcript discusses the integration of Natural Language Processing (NLP) and Generative AI in education and the workplace, emphasizing ethical and secure applications. The speaker introduces NLP as a field of AI that enables interaction between human language and computers, explaining its role in processing, analyzing, and understanding natural language. The discussion also highlights the ethical considerations in AI, including human-centric approaches, transparency, non-discrimination, and privacy. Furthermore, the transcript examines the role of AI in education and the workplace, noting its potential for personalized learning and automated tasks, while stressing the importance of governance and policy to manage these technologies. The text concludes with insights into AI's potential risks, like cybercrime and data poisoning, advocating for awareness and adherence to international standards and ethical guidelines.

      Highlights

      • ChatGPT reached 1 million users in just 5 days, highlighting the rapid growth of generative AI. 🚀
      • NLP helps machines understand human language structure and meaning through syntax and semantics. 💬
      • AI in the workplace can streamline tasks, improve productivity, and aid in decision-making. 📈
      • There are concerns about AI being used for misinformation and deepfakes, necessitating ethical considerations. ⚠️
      • Cybercrime can exploit AI for malicious purposes, emphasizing the need for robust security measures. 🛡️

      Key Takeaways

      • NLP enhances human-computer interaction by understanding and processing language effectively. 🧠
      • Generative AI, like ChatGPT, focuses on creating new content, such as text and images. 🎨
      • AI in education promotes personalized learning experiences and automates grading. 📚
      • Ethical AI integration should be human-centric, transparent, and fair. ⚖️
      • AI can boost workplace productivity but requires solid governance to prevent data leaks. 🔐
      • Being AI literate and ethically aware is crucial for adapting to future AI advancements. 🤖

      Overview

      In a rapidly evolving digital landscape, Natural Language Processing (NLP) stands as a pivotal tool in bridging the communication gap between humans and machines. By breaking down text into understandable components, NLP aids in comprehending complex language structures, enabling effective interaction. From sentiment analysis to language translation, NLP drives many applications we interact with daily, making it an indispensable aspect of modern AI.

        Generative AI, exemplified by tools like ChatGPT, takes the AI landscape a step further by crafting new content, be it text, audio, or images. While NLP focuses on understanding, generative AI specializes in creation, catering to the growing demand for innovative content generation. However, the fusion of AI into daily life, be it in education or workplaces, calls for a delicate balance of embracing technology while upholding ethical standards.

          With AI's increasing role in various sectors, ensuring ethical use becomes paramount. As AI becomes intertwined with day-to-day tasks, from automating workloads to enhancing learning experiences, the importance of human-centric design, transparency, fairness, and privacy cannot be overstated. The dialogue around AI's impact touches on both its vast potential and looming threats, like cybersecurity risks, requiring vigilant governance and a commitment to international ethical standards.

            Harnessing NLP: Ethical and Secure AI Integration in Education and Workplace Transcription

            • 00:00 - 00:30 [Music] okay so for today I'm going just to give a little introduction uh about the topic and then since um I was told to discuss about NLP I would also like to discuss a
            • 00:30 - 01:00 bit about the difference between NLP or natural language processing and the Gen AI or generative AI ethical use of AI or NLP efficient use again allow me to use it uh in two areas uh education and in the workplace no specific industry and when we say industry I won't talk about its application in the hospital or in banking or in um let's say oil and gas uh Etc so I'll choose education and work
            • 01:00 - 01:30 because I think uh some of the a good percentage of the attendees are students and the others are newly graduates as well and a future of NLP in the industry what is uh NLP NLP or the natural language processing it's a field of AI that focuses on the interaction between the human and human language and uh the computers so it involves processing and analyzing large amount of natural
            • 01:30 - 02:00 language such as data or such as text and speech and that it extracts the meaning and enable the communication between the humans and the Machine I will not be talking about very technical details about NLP I believe and I've mentioned this to Richard let the technical one be mentioned by the next uh be discussed by the next technical speaker so how does the NLP work and in n LP excuse
            • 02:00 - 02:30 me uh this involves breaking down a text into uh smaller components like words or phrases or sentences so it understands grammatically the the structure or there is a certain syntax and act actual meaning or the um the the semantics so the purpose of this is um it helps AI models accurately process the text so NLP you this is the mo uh is a mod model
            • 02:30 - 03:00 used to accurately process the text it identifies keywords the relationship between the meanings and it facilitates a better comprehension of the human language so even if you say let's say the the correct um sentence is my name is Irene Corpus even if you say name Irene Corpus it understands something like that okay so the in context of learning this reper refers to the nlp's
            • 03:00 - 03:30 capability to enhance its understanding by learning from past data or interaction or conversations it allows AI to interpret the context of the the sentences it adapt responses responses also based on interactions and it improves the accuracy and relevance over time so the the NLP is a a broader field it's a broader field of AI and it is focused again on interpretation of human
            • 03:30 - 04:00 language it can interprets it such that the machine and the humans gain um develop an understanding of what they are talking about it it converts one language to another even for that one it can do there's a speech recognition it converts broken words into text text to speech as well converting uh written texts into SPO uh spoken words um what are are examples of this um uh NLP tasks
            • 04:00 - 04:30 classification sentiment analysis so if you're going into some website you will see a small icon at the bottom right like the like a robot that has a uh a headset no this is the the chat bot so sometimes if you are interacting with the uh with the chat bot the way you enter let's say your text it can analyze
            • 04:30 - 05:00 whether you are happy you are sad you are frustrated this is what it says it means when you say sentiment uh analysis it can do language translation as well and summarization so the NLP can use both a statistical method and deep learning to process your language it makes uh making it a fundamental uh for a range of applications including grammarly and the Google translate or translate.google.com and other search engines
            • 05:00 - 05:30 okay so what are these examples of NLP uh models you have the voice assistant like Alexa if anyone of you own the Alexa you know that when you say Alexa play the um Spotify and then it will play your Spotify there's also the Siri or the Google Assistant sometimes even if you just put your phone near you and you speak there is a Quee wherein the this uh Voice assistance will initiate the language transl
            • 05:30 - 06:00 tool you have a Microsoft translator and you also have a Google translate it's translate.google.com no at the a very early time when the Google translate was implemented or launched it doesn't have it it creates a poor translation sometimes as since I'm I'm uh Comm I communicate with um uh various offices globally and I see a communication that is in different languages I copy and paste it in the Google translate but when it translates
            • 06:00 - 06:30 it it's like you won't understand anything but right now um it is very high in terms of accuracy of translation um you can read you can understand the context of the discussion because it creates a better uh translation at this time uh especially here in the UAE I get some Communications that are written in Arabic not only in Arabic and then English text but Arabic language itself when I paste it in the Google Now I'm able to uh to understand the content
            • 06:30 - 07:00 another um NLP tool is the text analytics I've um I'm not sure whether you have heard of the IBM Watson um the IBM Watson um there's also uh AWS comprehend and the Microsoft assure texts uh analytics and uh there is the um sentiment anal analysis API or um
            • 07:00 - 07:30 application program interface like the monkey learn or alien chatbots I've mentioned earlier not only in some but there are a lot of applications right now or web uh websites instead of having a real human to interact with the customers now they Implement an AI chatbot this um allows the company to have a simplified communication
            • 07:30 - 08:00 sometimes we see it as a simplified communication because there are already predefined questions or the FAQs and the chat bot will um reply according to the predefined question even the question sometimes you will see how can I help you today and then there are already options of the questions if you click on that there is already a predefined uh answer however if your answer is not your question is not satisfied in the end there hopefully not all of them but some of them will
            • 08:00 - 08:30 still say option give an option to talk to a um to talk to a human and that is the time where you will be engaging with the humans so these are the V virtual agents and other NLP tools the grammar and spell spell checker I use grammarly when I am creating um formal emails or when I'm documenting let's say policies part of my work is policy cyber policy development so before uh before I submit
            • 08:30 - 09:00 it for review I make sure that it is properly uh written uh grammarly correct even the punctuation the comma the semicolon these are very important when you are creating a document that will be published in public and be used uh by the public and in fact even the grammar Lee now can be integrated with your msword or your uh word processor no uh there is an option also even without the
            • 09:00 - 09:30 grammarly in Ms word there is an option to review the grammar uh or spell checker so that's is for NLP but before I move forward there I wanted to put something that relates to GPT because they are similar but there is still a difference but there is a difference between NLP or natural language processing and the GPT now GPT let's look at the Sprints to 1 million users
            • 09:30 - 10:00 Netflix took 3.5 years to reach 1 million there's also the kickstarter 2.5 and Twitter two years before it reached 1 million users the Facebook it took 10 months to reach 1 million users Spotify it reached five months to reach 1 million inst gram 2.5 months
            • 10:00 - 10:30 okay so it took five days only when chat GPT was launch officially launch to reach 1 million users Okay now what's the difference between NLP and generative AI chat GPT is just one of the generative AIS NLP interaction between human language and computers so it involves processing analyzing large amounts of natural language data such as
            • 10:30 - 11:00 text and speech and extract the meaning and enable communication between the humans and the machines no the generative AI on the other hand it the focus is that it specifically creates a new content including text images or audio and code it emphasizes um on producing new outputs rather than just understanding or analyze the existing input so functionally it typically uses deep
            • 11:00 - 11:30 learning like Transformers the GPT so whoever mentioned generative pre-trained Transformer then that's what it is a chat GPT is a a gen AI it generates coherent and um contextually relevant text there are there is also um model uh another de learning model like the diffusion model
            • 11:30 - 12:00 uh it is used in generating images and other visual uh media um for those of you who are using the chat GPT you will see on the upper left side that if you click one icon or button there it will show you other um gpts that are integrated with the open AI uh you can see D doll e for image and there's also Juke um jukebox I think J
            • 12:00 - 12:30 Jun box whatever for music so there's the Gan or generative adversarial Network this is used for image syn synthesis video creation and other types of creative content Let's uh talk about what else the fake the fake technology so how it works it uses the gun by the way Gan for deep fake it creates highly realistic uh uh synthetic images or
            • 12:30 - 13:00 videos of people make it making it appear as if someone is saying or doing something that they haven't again make yourself um aware or educated read and listen to the news not only in the country not only in the Philippines in the region Asia region and globally you may have seen a lot of deep fake that has been uh done with uh let's say President Obama former
            • 13:00 - 13:30 president Obama um Bill Gates um a lot of other celebrities the Gen the this deep fake are use are using the gun or generative adversarial Network in the gun there is the discriminator the generator and the discriminator so the generator this is the one that creates the fake images or or videos of the person while the
            • 13:30 - 14:00 discriminator this evaluates the creation this creation to distinguish them from the real vide so the two models they comp they compete and improve uh over time and unfortunately it's resulting in increasingly realistic dip fakes um let's see the next one So Co covering this very important for us to know is that there are ethics
            • 14:00 - 14:30 and principles in Ai and these are ethical consideration in AI integration um there are many companies now who are adopting and implementing the AI integrating them in their uh in their day-to-day work there is there was the recent gitex it's a one week technology event and a lot of applications AI applications have been launch there mostly by um either private or
            • 14:30 - 15:00 government entities and some of these are let's say procurement applications really designed for them and uh HR applications so this HR applications they already embed the information from the employees it analyzes um the the necessary training required for the employee especially after the performance appraisal at the end of the year it should come out with the the training that would fill in the
            • 15:00 - 15:30 Gap in skills and knowledge of the individuals also that is based on the objectives and let's say the job description that's written for the employees so some of the ethical considerations when you do AI integration are um human Centric um although AI has the potential to significantly enhance decision or human decision uh AI becomes more integrated
            • 15:30 - 16:00 into our daily lives it is crucial to ensure that it is designed and implemented but aligned with this key principle um in h being in human a human Centric AI or integrated with human Centric principle in mind and let's give an example an AI assistant designed to Aid doctors by providing suggestions they provide suggestions not make the decisions it still keeps the control
            • 16:00 - 16:30 in human hands so it it ensures this human Centric IT addresses um that the AI complements the human roles only enhancing the safety and decision making rather than replacing the human judgment transparency and explainability uh what can we give here as an example a financial AI tool for example it shows users what why a loan was approved or
            • 16:30 - 17:00 denied explaining the criteria and the logic for for this uh decision so it helps users understand an AI decision increasing the trust and enabling also informed decision making so if you notice decision making still lies into the hands of the humans fairness and non-discrimination this is what's very important for me an AI for for example hiring tool I
            • 17:00 - 17:30 mentioned about uh HR earlier so an H an AI hiring tool um audit audited to ensure it doesn't favor any gender so this is an example it doesn't favor any gender any race or background in candidate selection very familiar T that if you submit your CV it has your picture to me I don't need to put my picture because at this time you should know that
            • 17:30 - 18:00 whether it is AI or whether it is human your looks or your gender or your your ra race or your nationality even your religion should not be the basis for you to be hired in a position the same thing for AI it should be fair and non-discriminative it prevents biases it should prevent biases that could lead to unfair treatment or discriminatory outcomes privacy uh and privacy
            • 18:00 - 18:30 protection uh safeguarding privacy and ensuring robust data governance practices example uh a health care chatbot uh it anom anonymizes user information to prevent misuse of personnel Health Data uh for this one IT addresses the protection of user data from unauthorized access and ensure compliance with data protection regulation in EU we have the GDP in EU
            • 18:30 - 19:00 there's the gdpr general data protection regulation in the Philippines there's also the data protection uh law or policy um it was given or mentioned at the opening of the uh of this webinar accountability and oversight so uh this me means clear accountability mechanism for AI advocating for responsible a uh entities to be identifiable and held accountable so example is an AI surveillance system
            • 19:00 - 19:30 with the human review process to Monitor and verify any flagged Behavior or alerts IT addresses um the um or ensures that responsibility for AIS action allowing for correction of errors and accountability in deployment this one if I may just in um inject something something that I read just a few just an hour before the webinar started there's a 14-year-old uh teenager in the US who
            • 19:30 - 20:00 committed suicide and the lawyer or the parents now are suing the platform because um it turned out that the 14-year-old started engaging or interacting with a chatbot so the the teenager is a boy and started interacting with a chatbot who happens to be a character of a female and then cut it short he fell love with the
            • 20:00 - 20:30 chatbot and then you can just Google that please uh can AI be blamed for the suicide of 14-year-old try to read the U uh the the news it will show there that the last chat that he had with the chatbot is that I hope I can come to you now and the chatbot said I'm waiting for you if you can come now something like that okay try to Google it fresh news so AI in general so there are distinct
            • 20:30 - 21:00 considerations reason why I talk about nle first and then the GPT is to determine the distinction between the two AI um uh generative AI again is there's a Content authenticity generative AI um has a concern about misinformation deep fake or fake creation the in terms of usage restriction there must be Safeguard to
            • 21:00 - 21:30 prevent the use of gen AI in creating or uh creating misleading or harmful content that's why the uh the responsibility or R AI responsible Ai and not only the use but also in development is very important on the other hand the NLP the consideration there is the accuracy in interpretation NLP needs to be accurate and the and context aware to avoid misinterpretation especially in
            • 21:30 - 22:00 critical applications like legal or medical text analysis I um I recently actually just got out from the hospital and when I was reading all of the results especially the uh the blood test the ultrasound and the other tests it mentions at the end that the result is generated through voice recognitions so when when the doctor was doing let's say
            • 22:00 - 22:30 an ultrasound or a city scan it has a microphone it talks it speaks and interprets what he or she was saying during the process of the ultrasound or the city scan and that gets translated to a medical report but at the end it will say that that was generated by uh um a text a transcription or a voice transcription it is very important and then that the doctor will have to read and evaluate
            • 22:30 - 23:00 the content of that medical report now let's look at secure AI integration in a in uh education and workspace when we say secure AI integration and education in education and the workspace it depends on how you are implementing the AI so do you implement it um coming from the platform itself or you're going to implement it in your own platform or in your own uh
            • 23:00 - 23:30 environment AI in application in education uh can um for you for the students AI can assist in um providing personalized learning experiences for the uh for the faculty it's automatic grading and also improves actually um student engagement however AI must be integrated securely to minimize the risk so AI application in the workspace can
            • 23:30 - 24:00 automate on the other hand so education let's say in the workplace it can automate routine tasks improve productivity and even identify patterns in data in cyber security we have those of you who are working in it it operations or security operations there's what we call sock Security operation Center most of the tools that we are that are being used in there are are already a AI enabled like the seims
            • 24:00 - 24:30 uh security incidents and events management it already correlates data and gives you a meaning meaningful Insight if there is anything happening in the environment it will tell or mention that there is something happening um therefore it is only now the the uh the security officer's decision to look and analyze that data but definitely the bigger part of data analysis and the patterns is already done by the by the AI so this improves
            • 24:30 - 25:00 productivity and more um more efficient in addressing uh situation AI also in workplace has to have a corresponding AI use policy if any one of you here company are implementing AI make sure that you have a governance and policy in place in uh last year in April Samsung Electronics there was an employe who
            • 25:00 - 25:30 inadvertently leaked sensitive company information by using chat GPT and we know chat GPT uh it still continuously improves no if there's anything that is let's say unus unusual and reported by the end users they listen to that and uh one of the things I heard as well in the chat GPT is someone complained that why is somebody else's discussion or communication coming out in my GPT you're checking in or logging in when
            • 25:30 - 26:00 you're using chat jpt so you are expecting that most Communications are coming from your previous entries as well but at that uh at a certain point somebody complained that or raised a concern that somebody else's discussion is coming into his GPT so that case in the Samsung the there was an engineer who input who input a confidential data including source code internal meeting notes into the AI tool um it led to the
            • 26:00 - 26:30 information being stored in the external server so in response the Samsung prohibited the use of gen AI like the chat GPT on company devices and network to prevent data laks so again very important to when you implement AI make sure that there is a governance and policy in place if you are using an AI make sure from time to time to check on those uh ethics um ethics and principles
            • 26:30 - 27:00 that I mentioned earlier transparency accountability non-bias um human Centric so let's look in education how is it applied personalized learning uh the adoption of AI and Ed and education really raise questions about the future role of universities and faculty but here's how um how AI application complement rather than replace the traditional education structures question there is will we
            • 27:00 - 27:30 still need universities or faculty there are students here right there are faculty in here so will we still need University or faculty personalized learning in AI enhances the learning experience it adapts the content to individual student needs and learning styles so what is the role of the faculty I'm also an ajun faculty so I'm saying will I will they still need me so teachers and professors still play a
            • 27:30 - 28:00 crucial role in The Guiding mentoring and facilitating a deeper understanding and critical thinking and creativity no uh elements that AI alone cannot achieve in terms of automated grading a AI can automated grading for routine assignments saving the uh The Faculty time to focus more on complex student support so it's more on the personal relationship and support and mentoring to the student and the the grading is um
            • 28:00 - 28:30 repetitive task something that can be given to an AI but make sure for those students when you learn that your grade is now being automated through AI make sure that the faculty and I also um call the attention of the faculty the decision is still yours and please take care of reviewing the results of the AI now the RO Ro of the faculty more again more on human oversight to ensure the
            • 28:30 - 29:00 grating accuracy the fairness and to provide a personalized feedback especially for subjective uh assignments no AI in the workplace it's the um automation of routine tasks like um here in the zoom there's already the a automatic AI uh let's say Note Taker it creates the transcripts and um it can actually to some application the slack
            • 29:00 - 29:30 it can identify the to-dos or action items even identifies can depending on the workflow integration it can even identify who are the action owners and the due dates so it will create a notification that a certain uh action item discussed and agreed on the meeting is already due or is assigned to an individual so um what are the opportunities for staff in an AI enhanced uh workplace there are new AI
            • 29:30 - 30:00 driven roles AI integration um creates role like AI trainers or data analysts or AI Implement uh implementation Specialists or even ethics officers so huge or bigger organizations when they Implement AI they also integrate new roles like this in their organization imagine ethics officers is a part of um an organization now it's a
            • 30:00 - 30:30 new role when there is a heavy implementation of an AI this is where you will see how they try to um take it seriously especially in implementing the ethics and principles so the the focus now on the individuals or the staff is on strategic and creative work but uh AI enabled decision support is there however the final decision will still uh
            • 30:30 - 31:00 lie on the uh individuals the staff most especially the leadership um skills development and upskilling this is very important let let's forget the the our fear before maybe five years ago or even more that AI or the robots will replace us only those who don't know how to use the AI will be replaced now there are uh requirements in some um just try looking
            • 31:00 - 31:30 at job opportunities on LinkedIn for example you will find skills and um cap skills and knowledge that were not there before even for marketing and social media try to look at the requirements what are these requirements that were not there before imagine um sometimes they will they are looking for social media influencer that is now a new skill that can be uh integrated by
            • 31:30 - 32:00 a company when they are looking for social for marketing and sales um individuals uh even the use of Tik Tok no depending on depending on the um depending on the uh position or the job that you are applying for look at the skills that are looking for uh enabled uh what do you call this companion working together with an AI um new graduates my advice to you as
            • 32:00 - 32:30 well try to look at use your Linkin because when you apply for a job the L the a the Linkin is already has an AI enabled um uh classification or qualifying you already how many of your skills meet the skill requirement of the job that is being posted and from there you will know what are the skills that you need to to upskill or whether you need to rescale so preparing for the future to
            • 32:30 - 33:00 effectively prepare for the future with AI individuals need to focus on this area develop AI literacy um it involves continuous learning about the new AI development the tools and best practices which will help the uh you as individual to stay informed and capable of using AI effectively in your role so very important it is not asking you to to trans uh let's say move to another role
            • 33:00 - 33:30 but stay informed and how AI can help you in your or be effective in your role the ethical awareness it is equally uh critical staying updated with the AI ethics the regulations guidelines and ensure that AI is used responsibly protecting your use you as a user and your organization from potential risks and lastly uh adaptability it is essential for integrating emerging AI
            • 33:30 - 34:00 tools and approaches we cannot get away with AI anymore and in one conference in one uh last week in the gitex I was one of the panelists and one of the audience asked me how can we stop Ai and I said why will you stop AI right and I uh he said that there are his point of view and highly respected that is there are now a lot of he he feels that AI um integration is creating
            • 34:00 - 34:30 some not more okay some chaos and it is a boo um what you call a detrimental rather than beneficial so I asked him did you stop smartphones when you were still using an noia 3210 okay so were you able to stop the evolution of smartphone at that time and and did you ask if you can stop the
            • 34:30 - 35:00 development of further or a smartphone or further enhancing smartphone so AI is there you have to embrace it and um you have to live with it again uh I also do Consulting and at one point I love giving opportunities I love mentoring at one point I have to take someone in the Philippines for a job for a task um just also to give an uh an experience
            • 35:00 - 35:30 of how it is working in the UAE but remotely and my question is how good are you in using chat GPT okay and uh I said use the chat GPT because I know he doesn't have the right level of skills but using the chat GPT will help him learn what is the subject he has a certain idea of the subject check the task but coming out of the uh the
            • 35:30 - 36:00 outcome itself I would still prefer someone who knows how to use GPT if you don't know it use the GPT you will learn from it and you will increase your knowledge from it so don't don't say that to me if anyone says H CH GPT now I'm not using my brains no you use your brains together with the GPT okay I we still say that in the end we need our brains to work we need our brains to
            • 36:00 - 36:30 work with the AI and lastly uh again I cannot get away with this slide because I'm in I am in cyber security remember that there are cyber crime in AI cyber crime in AI highlighting how AI can be exploited by malicious actors Aid driven social engineering before we can see we can determine that there is a malicious email now with AI you will barely notice it uh
            • 36:30 - 37:00 automated attacks for us we are very careful about that because AI can automate cyber attacks like credential stuffing or distributed denial of Services Network scanning you know making them faster and more difficult to uh to to detect there are defects for fraud AI generated defects can be used for financial fraud and in fact there was um a it was last February or March that uh a
            • 37:00 - 37:30 huge uh company in the UK W lost 25 million I think double digit in millions because they approved the release of such fund and they thought it was legit because the the um the adversary even did a zoom meeting with them and unfortunately it was a deep fake CEO AI generated malware the AI helps
            • 37:30 - 38:00 attackers identify high value Target they can predict vulnerabilities and adopt malware tactics by bypassing the defenses um AI generated malware can also create more sophisticated malware capable of learning from detection pattern soia already if I am the AI generated malware I already know that this is how your AI can detect me therefore I can also evade your security
            • 38:00 - 38:30 measures more effectively uh gathering information the Cyber criminals use the AI to to scrape massive amounts of data from public and private sources um also reason why I mentioned earlier very important be be aware about the uh how your data is being processed whatever in whatever reason uh you're using an AI and uh data poisoning so attackers can
            • 38:30 - 39:00 feel feed false data into your AI system to manipulate the outcome or degrade the performance so in um conclusion there are potential benefits but there is an importance of uh being aware of the security as well um ethical considerations we have discussed that there are other principle AIS ethics and principles be aware of um
            • 39:00 - 39:30 other policies globally so you are not not only informed uh of what is happening within the country but be uh let's say um encouraged to learn as well how the other countries are implementing it and how you can apply it not maybe not in your own uh organization but at least even on a personal level sometimes we cannot demand more from the company you know this requires budget this requires big decision uh but if you can
            • 39:30 - 40:00 apply the the uh let's say grab the benefits of AI for you personally enhance your well-being and your profession and your career then do so uh make sure that you are um preventing an authorized access if you can um do not upload so much confidential and personal in information in any of the publicly open
            • 40:00 - 40:30 Ai and um yeah uh ethical considerations for for uh work space workplace and in education maintain fairness prevent biases uh adhere to privacy regulation and ethical guidance and in AI deployment and for those of you who also would like to know how to implement an AI governance put governance risk and compliance in your AI implementation
            • 40:30 - 41:00 there is already an AI management system ISO 4201 I'm in the uh process of take getting my ISO 420001 lead lead auditor and there's a lot of things to learn from this ISO management system