Tradespace’s Post

After two days of conference excitement at IP Service World, take the opportunity to explore expert insights in a very focused, hands-on environment with our exclusive workshops. Join our Post Conference Workshops! Today, we would like to introduce one of the three workshops: Hands-on AI Drafting and Prosecution Workshop In this workshop, participants will generate a patent application step-by-step, using advanced tools like machine vision to describe drawings and Qthena to review prior art and assess novelty. They will also draft novelty and inventive step arguments, with the option to bring their own cases for hands-on exploration of AI-assisted patent prosecution. Bring your laptop, get a free access and work on your real data! You may want to use published claims or public material for those drafting tests. ▶️ Bring Your Own Drafting Case: generate a patent application from claims or invention disclosure, step-by-step ▶️ Bring Your Own Drawings: use machine vision to pre-describe drawings, along with other advanced tools ▶️ Bring Your Own EP Case: Participants are encouraged to bring their own EP publication number (preferably a granted case) to explore how the Qthena solution can assist in generating novelty and inventive step arguments. ▶️ Efficient Prior Art Review: Using Qthena, participants will review the prior art in relation to claim 1 and receive an objective analysis from Qthena on whether claim 1 is novel in light of the provided prior art. ▶️ Drafting Arguments: The session will conclude by guiding participants through drafting novelty and inventive step arguments, leveraging Qthena‘s unique attorney-like skills to prepare arguments suitable for submission to the patent office. ▶️ Discuss and challenge: discuss the horizons of AI work in drafting and prosecution with the founders of qatent and Qthena Looking forward to an exciting AI day in Munich with you! François Veltz & Tony A. 📅 Workshops: 27th November 2024 📍 Location: Munich, Germany 🔗 Register now at https://lnkd.in/gKCtf38 and be part of IP Service World 2024! #IPServiceWorld #Innovation #Workshop #Munich #ipsw #Questel #IP #AIDrafting #Prosecution

Goodbye, patent hassles. Hello easy patent process! Discover how senseIP is transforming the patent process for inventors with AI patent automation. Seeing is believing! Includes: Ideation Freedom to Operate Research/ Landscaping Research Market Analysis Full patent drafting preparation & automation One click filing to the patent office Read all about it in our new blog: https://hubs.ly/Q02FCPB20

DeepIP (Formerly davinci)’s Copilot now assists patent practitioners in streamlining the preparation and response process for office actions! All you need to do is upload your context documents and the AI Copilot will help you: ✅ Summarize an invention, prior arts, and previous OA 💡Grasp the examiner’s evaluation ⭐ Identify key arguments and aspects of the OA ✍️ Suggest claim amendments And more… This improvement allows the attorneys to work better and faster while keeping control on what’s redacted. Super useful right? Need more reasons for attorneys to try this feature? Visit: https://lnkd.in/dwUaBu85

What if you had a PhD sidekick for all your patent work? OpenAI just released their new o1 model. Insiders speak of PhD-level reasoning capabilities. Unlike its predecessors, o1 can "think" before responding. This approach is said to allow it to tackle complex, multi-step problems and self-verify its answers. What does this mean for the patent attorney profession? Some thoughts: (1) o1 is said to have unforeseen capabilities to reason about complex mathematical and technical concepts. Could such a model be a human-level sidekick for drafting highly complex patents like for advanced AI algorithms? (2) While previous models would only identify direct similarities in the data, an advanced model like o1 might see more nuanced connections. Could this be a leap forward in AI-assisted patentability, infringement and freedom-to-operate assessments? (3) By simulating human-like reasoning, might the model be able to predict patent examiner objections, allowing patent attorneys to address these proactively in the application? Just my initial ideas. What are your thoughts?

🚀 Transform Your Patent Prosecution Workflows with Case Automations in Qthena by ipQuants AG! 🚀 Are you ready to revolutionize your patent practice? Introducing Case Automations – your new secret weapon for unparalleled efficiency and insight. 🔍 What’s New? Cross-Family Citation Tracker: Instantly discern interconnections of prior art references within a patent family, ensuring a robust understanding of the citation landscape. Patent Family Status Matrix: Get a holistic view of all family members and their current legal status, streamlining the assessment of a patent’s global stance. Prior Art Date Comparator: Directly compare your application's filing dates against the publication dates of cited prior art, making novelty and inventive step arguments a breeze. 💡 Why You’ll Love It: Save Time: Automate key examination workflows and reduce manual effort, allowing you to focus on what truly matters. Gain Insights: Access crucial insights into patent families, citation dynamics, and priority dates instantly. Cost-Effective: Enjoy pay-as-you-go activation and only incur costs for the automations you use. Qthena is here to elevate your patent examination process, providing you with the tools you need to make informed decisions faster. Ready to see Qthena in action? Schedule your demo today and discover how our new Case Automations can transform your practice! 🔗 Schedule Your Demo: https://lnkd.in/eWsw6MsJ #patents #AI #LegalTech

Are free, instant patentability searches the future? Inventors and Service Providers - I believe we're on the cusp of a major shift in how we conduct patentability searches. I envision that soon, the Patent Offices across the globe may offer free, instant search services. This sounds strange, but I have reasons to believe that this may happen in the future. Imagine submitting your invention disclosure and getting a detailed search report right away. It would be like using an ATM for patent information. You input your idea, and out comes your search results! Here's what this future might look like: 1. Government-run machine learning platform-driven free and accurate searches replace third-party providers. 2. AI and machine learning produce fast, accurate results; and 3. Search quality improves over time as systems learn. I say this because we're already seeing progress in this direction. The UK Intellectual Property Office has introduced 'SEARCH', an AI-driven tool based on European Patent Office technology. It helps examiners conduct more thorough and efficient searches. Presently its not accessible to the general public, but nothing stops them from doing so. For Inventors - this shift could lead to, faster innovation cycles and lower costs, For Patent Offices - this shift could lead to higher-quality patent grants. For Service Providers -  they may need to adapt and focus on analysis and strategy more. Thoughts?

Yohoo! Patent granted - Enhanced user selection for communication workflows using machine-learning techniques Patent number: 12001984 Abstract: The present disclosure relates to systems and methods for intelligently selecting users for inclusion in workflows. In some examples, machine-learning techniques can be executed to intelligently expand the set of user profiles included in a workflow. The intelligent selection of new user profiles may be continuously performed over time intervals, thereby enhancing the computational efficiency and accuracy of expanding the user profiles selected for inclusion in the workflow. Type: Grant Date of Patent: June 4, 2024

Contemplating the Impact of the USPTO’s AI-Focused Patent Eligibility Guidance on Networking Applications | Mintz - Intellectual Property Viewpoints - https://lnkd.in/eketGa5a - #AIFocused #Applications #Contemplating #Eligibility #Guidance #Impact #Intellectual #Mintz #Networking #Patent #Property #USPTOs #Viewpoints

💥💥💥 No Free Lunch in LLM Watermarking: Trade-offs in Watermarking Design Choices Advances in generative models have made it possible for AI-generated text, code, and images to mirror human-generated content in many applications. Watermarking, a technique that embeds information in the output of a model to verify its source, aims to mitigate the misuse of such AI-generated content. Current state-of-the-art watermarking schemes embed watermarks by slightly perturbing probabilities of the LLM’s output tokens, which can be detected via statistical testing during verification. Unfortunately, this work shows that common design choices in LLM watermarking schemes make the resulting systems surprisingly susceptible to watermark removal or spoofing attacks—leading to fundamental trade-offs in robustness, utility, and usability. To navigate these trade-offs, we rigorously study a set of simple yet effective attacks on common watermarking systems and propose guidelines and defenses for LLM watermarking in practice. 👉 https://lnkd.in/dC9MmGpZ #machinelearning

Here is my latest question for my fellow patent practitioners: What is your opinion of the latest examples of patent-eligible subject matter issued by the USPTO (link below)? Starting with Example 47 (Anomaly Detection) - In this example, the USPTO asserts that a claim (claim 1) directed to an application specific integrated circuit (ASIC) with circuits for an artificial neural network (ANN) is patent eligible. In contrast, a claim (claim 2) directed to a method of using an ANN (in a slightly more specific way) is not patent-eligible. Without getting into my personal views of what should be patent-eligible, I don't think a court will likely find that the ASIC would be patent-eligible under §101. My concern for this example is that practitioners will lean on the examples (such as Ex. 47) to get individual patents allowed, but these arguments will not change the enforceability of granted patents. Thus, I am concerned that the latest guidelines will further deepen the divide between what the USPTO and the courts deem to be patent-eligible. What do you think? Do you agree with the USPTO's analysis of Example 47? #iplaw #patent #patentattorney https://lnkd.in/gS_kjZST