Smartlogiq

We’re #hiring a new Wordpress Developer in Tunis. Apply today or share this post with your network.

We're hiring, checkout this post! 👏 https://lnkd.in/e8MrTECR

🚨If your AI does the crime, you'll do the time, warns DoJ 🚨 That’s a great headline from one of my favourite online tech news outlets The Register. AI Compliance is now even more important than ever, as it is fast developing arena and still not a global or regional standard. Interesting that the DOJ has started to develop their stance on Corporate AI use. No doubt other jurisdictions are looking into it and developing their position. Hopefully we'll end up with something that is consistent across the globe. The U.S. Department of Justice has updated its guidelines, putting AI under the compliance microscope because they can see the risks growing. Here are the top 5 takeaways every business leader needs to know: 1️⃣ Companies will be held accountable for AI's actions, even if unintended. 2️⃣ Compliance officers must now assess AI's potential impact on legal compliance. 3️⃣ The DOJ expects proactive measures to curb negative consequences of AI use. 4️⃣ Prosecutors will scrutinize safeguards against AI misuse, including by insiders. 5️⃣ Continuous monitoring and testing of AI applications is now expected. 🛡️ How to mitigate these risks? 1️⃣ Implement robust AI governance frameworks, AI monitoring and audits 2️⃣ Conduct regular risk assessments, establish clear accountability chains 3️⃣ Invest in AI ethics training 4️⃣ Maintain comprehensive documentation of AI decision-making processes 5️⃣ Create transparent reporting mechanisms to ensure their AI tools operate ethically and legally Remember, in the eyes of the DOJ (or the equivalent body in your jurisdiction), ignorance is no longer bliss when it comes to AI. The concern here is that most audit/compliance officers are not necessarily up to speed with the tech and its functionality, and their response (or push back) may not be as measured as it should be to support the Corporate's AI uptake or innovation drive. https://lnkd.in/e6U5UMww #AICompliance #CorporateGovernance #TechLaw #LeadershipInAI #aitransformation #ECCP #cto

🎉Happy Saudi National Day!🎉 To all our friends and colleagues in the Kingdom of Saudi Arabia, wishing you all a wonderful National Day celebration, recognizing the incredible strides Saudi Arabia has made in its Vision 2030 journey—transforming industries, fostering innovation, and becoming a hub for global business and culture. #ksa #saudinationalday #countrytransformation #happydays Smartlogiq

👽Evaluating LLMs, a more complex challenge than a needle in the haystack! A long post but bear with me... As the AI bubble slowly bursts, and we go back to reality and continue our consideration on how to use this amazing technology in our business endeavours to achieve the business impact we seek, we must be thinking about the first step, which LLM to use? In the course of looking at various LLMs, I wondered how to compare and make a judgement, empirical or otherwise, as to which LLM to choose for a given objective. If we were choosing a database platform (Oracle, MySQL, PostgreSQL) for instance we take various metrics to measure against, such as supportability, speed, performance, migration headaches and so on. On a more simpler decision tree, if you were choosing a mobile phone you’d look at the aesthetics and check performance results from the likes of AnTuTu and GeekBench, and lastly you’ll probably look at battery longevity. When doing this with LLMs it becomes even more complex and the method of selection even more elaborate, as you can get different results and explaining them is even more complex. So how do you go about selecting an LLM and understand their strengths, limitations, quality of output and potential biases, considering most of them are a black-box and you’d have to have an army of engineers to test and investigate. As LLMs are constantly evolving, your evaluation process will also have to evolve to take into consideration the new changes/features being introduced. Of course you can use the same data set (assuming it is wide and deep enough) and a set of problems (e.g. Q&A, classification, summation, prompt categories, prediction) that they need to tackle and apply it to various LLMS and measure time, cost and accuracy. But is that enough? Is that sufficient qualification to make an informed decision? How do you measure the level of understanding of a prompt and the quality of the output, especially when language itself has nuances and possible interpretation! There is definitely a need for LLM evaluation, and how to do this on an industrial level and maintain impartiality is also another dimension needing consideration. Would love to hear anyone’s approach or opinion as to how you’re approaching this challenge as more and more companies are looking to utilise Gen AI and LLMs in their business solutions. PS These two images were generated by two different platforms, an anecdotal example of evaluation! #llm #ai #genai #aibenchmark #llmbenchmark #digitaltransformation

🚀 AI Adoption: Are You Ahead of the Curve or Falling Behind? Various published statistics show 20%-30% of CEOs have already integrated AI into their operations, while 10%-25% plan immediate adoption. Where does your company stand? Browse part 1 of the Smartlogiq essential "best practices" for successful AI integration in our latest blog post. From crafting a clear strategy to prioritizing data quality, we've got you covered. Don't miss out on the AI revolution. Click to learn how to harness AI's potential effectively! #AIAdoption #BusinessStrategy #TechInnovation #DigitalTransformation https://lnkd.in/e3A8dYdh

Best of luck Leila Ben-Gacem with the Ultra Mirage El Djerid challenge, everyone at Smartlogiq is rooting for you. What a great cause to do it for as well 👍👏

We believe that #IT #outsourcing will expand more and more over the years. However, do you know the #pros and #cons of each type of outsourcing? Dive into this #carousel to explore the differences between #Nearshore, #Offshore and #Onshore Outsourcing. Choosing the right model depends on your specific business needs and labor market conditions. If you would like to learn more about IT outsourcing and Smartlogiq's services, contact us here on LinkedIn!

Even if #cloud computing continues to grow, #onpremise solutions remain a vital part of many organizations' #IT strategies. From controllable #security, to easier #compliance and cost control, to higher customizability, many companies are reluctant to move everything to the Cloud. At Smartlogiq, we help our clients navigate these choices to find the best architectural fit for their unique needs. What are your thoughts on the role of on-premise solutions? Are they still going to retain their role? Find out more in this carousel!

Mastering #Agile Software Development is more and more important in companies. Agile can enhance #projectmanagement, improve #team collaboration and deliver high-quality #software. It is based on some core principles, like the collaboration with customers, the iterative development, and the flexibility and adaptability to be open to changes throughout the development process. Discover in the 2nd page of this document some #bestpractices that you can apply in your company for better Agile #adoption. At Smartlogiq, we have professionals that are trained in Agile methodologies and can support clients that develop software in this way. #AgileDevelopment #SoftwareDevelopment