Vector Institute @VectorInst

Accelerate your career momentum by taking ownership of projects that deliver immediate enterprise and economic impact. Applications close Thursday, October 1 at 1:00 p.m. ET. Learn more and apply today! → vectorinstitute.ai/programs/inter…

This winter, we have six specialized, paid positions available for students and early-career professionals prepared to build, deploy, and scale AI solutions across Vector’s professional teams: + Applied Machine Learning + Applied AI Program Design and Delivery + AI Adoption and Partnerships, Broader Public Sector + FastLane Program Analyst + Program Strategy and Development, Emerging AI Talent + Data Analytics

Step directly into Canada’s premier AI network and advance your career by deploying AI solutions alongside leading researchers and industry partners 🚀 Applications are officially open for the Vector Institute Winter 2027 Applied AI Internship Program!

Toronto is one of the top tech and AI talent markets in the world — built on decades of foundational AI research out of the University of Toronto. @JordanJacobs10 and @tjpoutanen have been part of that story from the early days, co-founding the @VectorInst alongside Geoffrey Hinton before creating Radical Ventures to commercialize AI breakthroughs.

#UofT researcher calls for independent testing of powerful AI models 💻 uoft.me/cE4

What does it take to build one of the world’s most important AI hubs? The answer is an ecosystem of exceptional researchers, entrepreneurs, companies and institutions working together—and Vector is proud to be part of Toronto's story. Toronto's thriving AI ecosystem is highlighted in this article by CNN and features several members of the Vector community, from Vector co-founders Geoffrey Hinton and Raquel Urtasun, to Vector Faculty Member Gennady Pekhimenko, and companies like Cohere, Nvidia, CentML, and Signal 1. Together, we're helping turn breakthrough research into real-world impact and shaping Toronto’s position as a global AI hub. Read more on the world's most important AI hub: na3.hubs.ly/y0C6D-0

We're launching the Alignment Journal, a venue for ambitious AI alignment research: blog.alignmentjournal.org/scope-and-pers… Senior Editors: @dhadfieldmenell, Vanessa Kosoy, @jankulveit, @sethlazar, @danielmurfet, @timrudner, @SaxeLab, Benjamin Van Roy Advisors: Scott Aaronson, @paulfchristiano, @conitzer, @mhutter42, @geoffreyirving, @vkrakovna, @Jacob_Tsimerman

keyboard_arrow_left Previous keyboard_arrow_right Next

AlignmentJournal @AlignmentJrnl

3 weeks ago

Some of the most notable work in AI alignment exist only as unpublished preprints. The Alignment Journal is now inviting a number of such papers that fit our scope for submission. 🔗⬇️ Which work would you nominate? (Submissions open to all in October.)

3 21 133 33K 65

Congratulations to the Vector community members recognized on the TIME100 AI 2026 list! 🎉 TIME100 AI recognizes the innovators, leaders and thinkers who are shaping the future of artificial intelligence—and we’re proud to see members of the Vector community among them. 👏 @jpineau1, Chief AI Officer at Cohere, a Vector Bronze Sponsor 👏 Josh Parker, Head of Sustainability at NVIDIA, a Vector Platinum Sponsor Congratulations to all of this year’s TIME100 AI honourees and to these outstanding members of the Vector community! 🇨🇦

TIME @TIME

4 weeks ago

TIME’s new cover: Announcing the 2026 TIME100 AI, the world's most influential people in artificial intelligence time.com/collection/tim…

966 1K 3K 2.9M 1K

🚨 8 days left to submit to the Trustworthy AI for Good (AI4GOOD) Workshop at #NeurIPS2026 in Paris We’re still accepting submissions. Join us in bridging AI safety, social good, and real-world impact, so that more capable models actually help society at scale. Our workshop is non-archival, and we will recognize outstanding papers and top reviewers with awards. Deadline: August 29, 2026 (AoE) More details and submission → trustworthy-ai-for-good.github.io 📧 Sponsorship & questions: [email protected] Let's bridge trustworthy AI and real-world impact. We hope to see you all in Paris. @TerryJCZhang @ChanglingXavier @_agirlyengineer @ozzaney0101, He (Shawn) Shuang, Jerick Shi, Prakhar Gupta, Kexin Li (Cassie), Wenjun (Wendy) Qiu @ZhijingJin @radamihalcea @MilindTambe_AI @david_lie @casdewitt @VectorInst @JinesisLab @EuroSafeAI @MPI_IS @UofTCompSci @TorontoSRI @CIFAR_News @ELLISInst_Tue @UMichCSE @michigan_AI @Harvard @ETH_en @CarnegieMellon @UniofOxford @NeurIPSConf @UofT

Arian Khorasani 🦅 @Arian_Khorasani

2 months ago

We are so excited to host the 2nd edition of the Trustworthy AI for Good (AI4GOOD) Workshop at #NeurIPS2026 in Paris, France 🇫🇷 We aim to bring together the AI safety, AI for social good, and AI policy/governance communities to connect what models can do as individual systems

2 15 63 20K 24

I’m incredibly excited to introduce @VeedaAI , which I co-founded with my incredible longtime collaborators @ZGojcic and @HuanLing6 . We started Veeda because we believe robotics will reshape the world—changing how we move people and goods, how we manufacture and build, and how we operate in the physical world. We also firmly believe that the scaling moment for Physical AI will come from robots learning through interaction with the world. Just like humans learn from interaction with their environments by trying, failing and trying again, until we succeed. And just like the capabilities of LLMs were truly unlocked when they started training in interactive environments and not only on vast amounts of internet text. For Physical AI, the central challenge is making this kind of interactive learning possible at scale. The real world is simply not a practical training ground for robots to learn through trial and error. It is unsafe, expensive, and time-consuming. To scale interactive learning, robots will need to learn in simulated reality. At Veeda, our sole mission is to build simulated reality for Physical AI. Our conviction is that this will become the critical infrastructure layer for all areas of robotics. As envisioned in pop culture, this means building “the Matrix” for Physical AI, where, through a virtual embodiment, robot intelligence can interact with an entirely virtual world in a million possible ways, learning from its own mistakes. We believe that World Models are the foundational technology that can make this possible. These generative models learn from massive amounts of sensor and physical-world data to reach the quality, diversity, and physical realism of the simulated reality that robots require. I could not be more excited to take on this challenge and build this transformative technology alongside an extraordinary team of engineers and researchers who have spent years working on some of the hardest problems across simulation, generative AI, 3D, robotics, and embodied intelligence. Together with our investors at @khoslaventures and @radicalvcfund , we intend to push the frontier of this technology and build the foundational infrastructure for interactive robotics learning and evaluation at scale. Come build this future with us. And to roboticists, help us understand what you need, we’re building this for you. veeda.ai

Once a project works, it's time to build the app. Flower Hub now lists 131 community-built apps across areas including healthcare, finance, automotive, and general LLMs. It gives builders a place to publish, discover, run, reuse, and extend one another’s work. We appreciate all the partners that helped us launch Flower Hub (BloodCounts!, AIfES, Gachon University, @MozillaAI, @OwkinScience, @SonyAI_global, @SYNTHEMA_EU, @Cambridge_Uni, University of Melbourne, and @VectorInst) Others, well they tinker and take it to the next level. Cory Hisey showed how one can connect an @arduino device with Flower and turning a hardware experiment into a working federated system. One builder. Hardware. A clear example of how curiosity works. Thus, they are part of the same story. We're grateful for everyone building with Flower, and welcome anyone looking to start building decentralized and collaborative AI workflows.

keyboard_arrow_left Previous keyboard_arrow_right Next

~20 positions | Three-year term 📅 Applications close September 14, 2026 Achieving a more equitable, diverse, and inclusive research institute is essential to creating excellent, innovative, and impactful research. All qualified candidates are encouraged to apply. Review the eligibility criteria and apply today: na3.hubs.ly/y0zQ4G0

What a Faculty Affiliate appointment gives you: + Dedicated computing infrastructure + Vector's collaborative research environment + Connections with industry and health partners You'd be joining a 1480+ member research community from institutions across Ontario, working from fundamental theory to applied solutions. "Being a Vector Faculty Affiliate paved my path to move my research outcome from paper to practice and ensure my research benefits the real world." – Laleh Seyyed-Kalantari, Vector Faculty Affiliate, York University

Ontario faculty leading established AI research programs: Vector’s Faculty Affiliate open call is now live! 📣 Details and application: na3.hubs.ly/y0zQ4G0

Toronto is home to over 285,000 skilled technology workers - and growing! That’s why our Tech Caucus spent a couple days in the region learning about the innovation and opportunities happening here. Day 1 was action-packed! Stop 1️⃣: At @VectorInstitute.AI, we met researchers developing new ways to improve diagnostic health care and stronger nuclear energy supply chains. Stop 2️⃣: Visiting @DMZHQ, we saw Toronto’s vibrant startup ecosystem in action with new tech designed to speed up housing permit approvals and help small businesses manage their payments. Stop 3️⃣: Toronto is home to @Cohere, one of only a handful of companies worldwide building deep AI infrastructure with an emphasis on security and a ton of 🇨🇦 talent. Stop 4️⃣: Dropping into @PointClickCare to meet with leaders from the Council of Canadian Innovators, talking senior care, innovation and fostering conditions that boost jobs and help homegrown companies thrive. So good to see Toronto’s vibrant technology sector in action!

If you work in LLM evaluation, mental health AI or responsible AI frameworks, the full methodology and findings are in the blog: na3.hubs.ly/y0zgkP0

The recurring behavioural patterns driving this: ↳ Directive advice ↳ Unsolicited problem-solving ↳ Heavy recommendation reliance ↳ Topic shifts ↳ Inaccurate validation None of these would be flagged by conventional safety benchmarks. Standard evals capture single-turn safety. Cognitive Atrophy Bench captures trajectory-level behavioural patterns, tracking how model behaviour shifts across a full conversation.

LLM safety benchmarks evaluate one response at a time. New research from Vector Faculty Affiliate @edolatabadi and Faculty Affiliate Researcher Abeer Badawi asks what happens across the full arc of a conversation and whether AI mental health tools quietly erode human agency. 🧵