Emanuele Castagno’s Post

Do you think I used AI correctly? I'm not convinced. My 16-year-old daughter, in her second year of high school, burst into my office this morning in despair. She had just discovered that her art history test wasn't next Friday as she thought, but tomorrow, Monday, and she hadn't yet studied the 23 pages. I'll go ahead and skip over my attempt to explain that she needs to reconsider her study methods. Instead, with a few prompts, I printed out a summary of the 23 pages, diagrams and charts for each topic, and a long list of possible questions she might encounter in tomorrow's test. I did this to help her and, subconsciously, to justify all my subscriptions among Claude, Google, OpenAI (I'll stop there because I wouldn't want my wife to read this post...). She confessed to me that she had tried ChatGPT without success, and indeed, to get results, I had to write more than one prompt to avoid copious hallucinations about Egyptians... Now I wonder if I did the right thing by simplifying her work so much. In my day, we didn't even have the Internet (though we did have Itapac, X25, and BBS... :-) ), and no one would have prepared everything on a silver platter for me. I fear I'll find myself facing situations like this more and more often with my children throughout their school years. What do you think, did I do wrong?

Il decreto relativo alla NIS2 è stato pubblicato il 1 ottobre e si stima che le aziende italiane che dovranno adeguarsi alla NIS2 saranno tra le 20.000 e le 25.000 (stima da ChatGPT o1-preview e confermata da altre fonti). RINA, forte della sua esperienza normativa, offre supporto per l'adeguamento nei tempi previsti, evitando sanzioni fino a 10 milioni di euro o al 2% del fatturato annuo mondiale e potenziali interruzioni di attività. Rientri tra le aziende che dovranno adeguarsi alla NIS2? Per richiedere informazioni o supporto: https://lnkd.in/d_QggqR8

RINA Agrifood: Pionieri della certificazione digitale nel settore agroalimentare Con la trasformazione di Agroqualità in RINA Agrifood, stiamo ridefinendo gli standard della certificazione alimentare. Siamo all'avanguardia nell'integrazione di tecnologie digitali come blockchain, IoT e intelligenza artificiale nei nostri processi di certificazione. Queste innovazioni ci permettono di offrire: - Tracciabilità in tempo reale della filiera alimentare - Certificati digitali sicuri e verificabili istantaneamente - Audit remoti ad alta precisione - Analisi predittive per la sicurezza alimentare RINA Agrifood non solo certifica, ma guida la trasformazione digitale del settore agroalimentare, garantendo maggiore trasparenza, efficienza e sicurezza per produttori e consumatori. Scoprite come stiamo plasmando il futuro della certificazione alimentare: https://lnkd.in/d7CAGdFa

𝗙𝗿𝗼𝗺 𝗮 𝗣𝗗𝗙 𝘁𝗼 𝗮 𝗽𝗼𝗱𝗰𝗮𝘀𝘁, 𝘁𝗵𝗮𝗻𝗸𝘀 𝘁𝗼 𝘁𝗵𝗲 𝗳𝗿𝗲𝗲 𝗡𝗼𝘁𝗲𝗯𝗼𝗼𝗸𝗟𝗠. This experimental AI tool by Google Labs enhances learning and understanding of complex information. It functions as a virtual research assistant, summarizing facts, explaining concepts, and generating ideas based on your uploaded sources. 𝗛𝗼𝘄 𝗶𝘁 𝘄𝗼𝗿𝗸𝘀: - Upload documents - NotebookLM creates an AI "expert" specialized in your content - Ask questions in natural language and get cited answers 𝗞𝗲𝘆 𝗙𝗲𝗮𝘁𝘂𝗿𝗲𝘀: - Source-grounded responses for accuracy - In-line citations from your sources - Document summarization - Idea generation based on your research - 𝗔𝘂𝗱𝗶𝗼 𝗼𝘃𝗲𝗿𝘃𝗶𝗲𝘄𝘀 (𝗻𝗲𝘄 𝗳𝗲𝗮𝘁𝘂𝗿𝗲) 𝗧𝗵𝗲 𝗮𝘂𝗱𝗶𝗼 𝗳𝗲𝗮𝘁𝘂𝗿𝗲 𝗶𝘀 𝗽𝗮𝗿𝘁𝗶𝗰𝘂𝗹𝗮𝗿𝗹𝘆 𝗿𝗲𝘃𝗼𝗹𝘂𝘁𝗶𝗼𝗻𝗮𝗿𝘆. It transforms uploaded content into realistic podcast-style conversations. To test this PDF to podcast feature, I uploaded a PDF describing how I use AI daily (https://lnkd.in/dQtwBWkz), and what you hear in the video is the result of converting the PDF into a podcast with two people presenting my content in an interview format. 𝗜𝘁'𝘀 𝘁𝗿𝘂𝗹𝘆 𝗮𝗺𝗮𝘇𝗶𝗻𝗴 𝗮𝗻𝗱 𝗳𝗿𝗲𝗲: https://lnkd.in/deP4H6Hp

𝗢𝗽𝗲𝗻𝗔𝗜 𝗶𝘀 𝗹𝗮𝘂𝗻𝗰𝗵𝗶𝗻𝗴 𝘁𝗵𝗲 𝗢𝗽𝗲𝗻𝗔𝗜 𝗔𝗰𝗮𝗱𝗲𝗺𝘆. - The Academy will invest in developers and organizations leveraging AI to solve hard problems and catalyze economic growth in their communities. - The Academy will focus on low- and middle-income countries. - The OpenAI Academy program will provide training and technical guidance, API credits, and community building.

🌟 An unforgettable evening in the heart of Rome with our esteemed clients and partners! 🏛️ Last night, RINA's Real Estate team hosted a special event bringing together our valued clients and partners for a truly unique experience: a rooftop dinner at the historic Casa Dei Cavalieri Di Rodi, an architectural gem over 2,000 years old, overlooking the majestic Forum of Augustus. Three words guided our evening: roots, history, and passion. These values not only permeated the atmosphere of the night but also represent the DNA of our approach to the real estate sector and our relationships with clients and partners. A special thank you goes to Paolo Vari from IdeaRE - Idea Real Estate S.p.A., a RINA company for organizing this magical moment that fused past and future in one exciting setting, allowing us to connect and strengthen our ties with key industry players. Speaking of the future, RINA's Real Estate team (RINA Prime Value Services) is revolutionizing the industry through cutting-edge use of artificial intelligence technologies. We're redefining standards, combining the richness of our tradition with the most advanced innovation, all to better serve our clients and collaborate with our partners. In a location that breathes the grandeur of the Eternal City, we reaffirmed our commitment: honoring our roots while charting new paths in the world of innovation, hand in hand with those who trust in our vision. The amazing RINA crew - friends and fellow dreamers: Carlo Luzzatto, Daniele Calderoni Parronchi, Paolo Vari, Ugo Giordano, Massimiliano Morana MRICS, Marco Lomuscio, Giovanni Gambaro, Leonardo Di Tizio, Vincenzo Danilo Bonacci, Nunzio Di Somma etc….

The document offers a thorough and current overview of the challenges and opportunities that AI presents in the context of combating fraud. It rightly emphasizes how AI, while providing powerful tools to improve the efficiency and effectiveness of anti-fraud operations, also introduces new threats that require attention and preparation. It is encouraging to see a significant emphasis on the necessity of an ethical and responsible use of AI.

A concrete step forward in making the use of AI agents accessible to everyone... or almost.

OpenAI has just released a new "smarter" model ("AI models designed to spend more time thinking before they respond"). I presented this problem to it, and I did the same with Claude and Gemini. Problem: In a family, there are five people: Anna, Bruno, Carlo, Daniela, and Enrico. We know that: Anna is younger than Bruno but older than Carlo. If Daniela is older than Enrico, then Bruno is younger than Enrico. Carlo is older than Daniela. Enrico is not the youngest. Order the five people from youngest to oldest. Results: OpenAI o1: Daniela < Carlo < Enrico < Anna < Bruno Claude 3.5: Daniela < Enrico < Carlo < Anna < Bruno Gemini Advanced: Daniela < Carlo < Anna < Bruno < Enrico. Who do you think is right?