Sign in to view Oscar’s full profile
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
United States
Sign in to view Oscar’s full profile
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
10K followers
500+ connections
Sign in to view Oscar’s full profile
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
View mutual connections with Oscar
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
View mutual connections with Oscar
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
Sign in to view Oscar’s full profile
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
View Oscar’s full profile
-
See who you know in common
-
Get introduced
-
Contact Oscar directly
Other similar profiles
-
Kristina Vega
Kristina Vega
Cyber Money Laundering in Real Estate Investigations Corp
15K followersMiami Beach, FL -
Laura Teclemariam
Laura Teclemariam
University of California, Berkeley
10K followersSan Francisco Bay Area
Explore more posts
-
Yoel Roth
Match Group • 5K followers
One of the hardest challenges in online trust and safety is knowing whether an account is real… and doing so in a way that protects privacy, scales globally, and makes the experience seamless for authentic users. That’s why the launch of Face Check is such an important milestone for us at Tinder. This new feature uses facial liveness verification to help confirm users are real and match their profile photos, and prevent impersonation and fake accounts. The proof is in the data: We've seen a more than 60% reduction in exposure to bad actors, and a more than 40% decrease in reports, alongside measurable improvements in user perception of authenticity. Face Check is already live in Colombia, Canada, Australia, India, and several countries across Southeast Asia, and is now expanding across the U.S., with plans to roll out to more markets and more Match Group apps soon. Grateful to our Trust & Safety, Legal & Privacy, Product, and Engineering teams around the world for leading the way. Learn more via The Wall Street Journal: https://lnkd.in/gvmE8QYn #TrustAndSafety #OnlineSafety #Privacy #MatchGroup #Tinder
200
12 Comments -
Katrina Mulligan
OpenAI • 11K followers
SCSP and Gallup have joined forces to provide this essential report on American AI attitudes, which is required reading for AI practitioners. This report reinforces Edelman's findings last year -- that only a third of Americans trust AI "fully" or "somewhat." By contrast, trust in AI is overwhelmingly higher, beyond 70% in China and the rest of the developing world. That means America could win the race to AGI on the technology side, and lose the race to China on adoption. What do we do to remedy that? Drive adoption. The SCSP study shows that experience with AI matters. AI users are twice as likely to trust it 46% compared to non-users 23%. I always say that you don't get fit by reading about working out, and likewise you don't get smart on AI without using it. This conviction was behind OpenAI's decision to give ChatGPT to the federal workforce for $1. And I'm proud to say, we're seeing major growth in public sector adoption every week. https://lnkd.in/efUWFCQM
70
8 Comments -
Vijaya Kaza
Google • 8K followers
I’m often asked for feedback on startups focused on securing Agentic AI. While these targeted solutions have their place, agent security is far too complex and nuanced to be solved by any single product or silver bullet. Beyond existing infrastructure and model-related risks, agents add new risks, which I group into three broad categories: 1. Risks from attack surface expansion: Agentic systems require broad access to APIs, cloud infrastructure, databases, and code execution environments, increasing the attack surface. MCP, which standardizes how agents access tools, memory, and external context, introduces a new kind of attack surface in its own right. Since agents take on human tasks, they inherit identity challenges like authentication and access control, along with new ones such as being short-lived and lacking verifiable identities. 2. Risks from agent autonomy: By design, autonomous agents make decisions independently without human oversight. Lack of transparency into an agent's internal reasoning turns agentic systems into black boxes, making it difficult to predict or understand why a particular course of action was chosen.This can lead to unpredictable behavior, unsafe optimizations, and cascading failures, where a single hallucination or flawed inference can snowball across agents and make traceability difficult. 3. Risks that come from poorly defined objectives: When objectives or boundaries are poorly defined by humans, even a technically perfect agent can cause problems. Misunderstood instructions can lead to unsafe behaviors, buggy or insecure code. In practice, the biggest challenge for teams building agents is opening the black box and understanding how the agent thinks, so they can help it behave more consistently and course-correct as needed. This requires strong context engineering to shape inputs, prompts, and environments, rather than relying on third-party tools that face the same visibility issues. Additionally, custom, context-aware guardrails that are tightly integrated into the agent's core logic are needed to prevent undesirable outcomes. No external product can prevent an agent from doing the wrong thing simply because it misunderstood a vague instruction. That can only be prevented by proper design, rigorous testing, and extensive offline experimentation before deployment. Of course, that’s not to say third-party AI/agentic AI security solutions aren’t useful. Paired with traditional controls across infrastructure, data, and models, they can partially address the first category of risk. For example, AI agent authentication/authorization to manage the lifecycle and permissions of agentic identities, and granular permissions for tools are good use cases for agentic AI security solutions. Penetration testing is another highly productive use of external tools to detect unauthorized access, prompt and tool injection, data and secrets leakage. #innovation #technology #artificialintelligence #machinelearning #AI
231
22 Comments -
Amanda Storey
6K followers
🔈Yesterday was our *inaugural* EMEA Anti-Scams and Fraud Summit, hosted at our Google Safety Engineering Center in Dublin. 🇮🇪 A few highlights included: Announcing that Google.org is committing $5M to anti-scams research, awareness and training in select European countries. And sharing how we are using Google AI within the Global Signals Exchange, a project — in collaboration with the Global Anti-Scam Alliance (GASA) and DNS Research Federation — that is fast becoming a global clearinghouse for online scams and fraud bad actor signals. Thanks to everyone who worked on this and our partners and guests for contributing to the important discussions! 🤝 Read more about the work we’re doing to combat scams here ▶️https://lnkd.in/ez3Z9TJ9 #TrustAndSafety #GSEC #AntiScamsSummit #onlinesafety #GoogleAI
180
2 Comments -
Jay Lauf
Charter • 7K followers
At our summit last week Charter and Indeed hosted a roundtable breakfast conversation featuring Svenja Gudell, Chief Economist at Indeed, on the state of hiring, skills, and the real pace of AI adoption inside companies. Svenja shared insights and data from Indeed Hiring Lab's most recent Labor Market Overview report, which you can check out her: https://lnkd.in/e-AxBKVf The gathering was under Chatham House Rule, so I can't quote anyone, but some takeaways that stood out: 💡 The labor market has shifted from “over-heated” to “uncertain.” We’ve gone from two open jobs for every unemployed person to roughly one. The slowdown lands most heavily on early-career workers — fewer internships, fewer first roles, more risk-aversion from employers. 🤖 Despite the hype, AI adoption is still early. Only ~10% of employers say they’re using AI, and just 3% of job postings mention it. Headlines suggest a revolution; reality looks more like cautious experimentation. 🪜 “Don’t skip a generation.” If we allow entry-level roles to collapse while AI reshapes work, we risk losing the pipeline of future managers, team leads, and culture carriers. The work now is building on-ramps, apprenticeships, and learning loops — not waiting for perfect clarity. 🧠 Jobs won’t disappear — tasks will change. The most durable framing we heard: AI doesn’t replace roles, it reshapes responsibilities. The question isn’t “Which jobs go away?” but “How do we redesign how work gets done?” 🎯 Hire for the human skills the software can’t replace. Critical thinking. Clear communication. Collaboration. Empathy. Judgment. The tools handle the mechanical. Humans handle the meaningful. 🏃♀️ The organizations that thrive will be the ones that learn fastest. Not the ones with the most data or the flashiest models — but the ones that create psychological safety to experiment, reflect, and adapt together. One line that captured the room: "We are still early. The work now is learning, not landing." Thank you Paula Judge, Ryan Smith, Amy Lui Abel, Jolen Anderson, Kevin Delaney, Serina Pak, Mahvish Zaman, Kit Krugman, Anne-Claire Berg, Lauren Uranker, Amanda Bhalla, Katie Burke, Joshua Horenstein, Christine Murphy O’Keefe, Lisa Moore (she/her), Brian Elliott, Isabel Berwick for joining and sharing your insights! Victoria Chen Norland, Jennifer Oliveras
55
16 Comments -
Elina Cadouri
Dock Labs • 3K followers
At WWDC25, Apple made a major announcement for digital ID adoption in the US: passport support is coming to the Apple Wallet in iOS 26. While it won’t replace your physical passport for international travel (yet), this feature opens up powerful new use cases for domestic air travel and digital identity verification. Until now, Apple’s support for digital IDs and mobile driver’s licenses (mDLs) has been limited to just 12 states, a slow rollout heavily dependent on state-by-state integration. However, by leveraging federal identity documents like passports, Apple has discovered a clever shortcut: instead of waiting for all 50 states to catch up, they’re leveraging the passport as a form of widely accepted, government-issued ID that can be digitized and used across a broad range of services. I suspect this will follow a similar model to Google’s implementation, where the digital ID in your wallet isn’t your actual passport, but rather a secure, cryptographically signed digital representation of your passport data. What makes this especially exciting is Apple’s support for the W3C Digital Credentials API. If that integration plays out as expected, it means users may soon be able to use their digital passport to verify their identity online with a single tap. This could be the spark the US digital ID ecosystem needs. By turning passports into reusable digital credentials and placing them directly into the most widely used smartphone in the country, Apple may give the entire market a serious boost. – Image: mDL presentation / Apple
53
9 Comments -
Carolyn Healey
Service Marketing • 11K followers
Company-wide AI governance sets the guardrails. Department governance runs the operating system. You need both to scale AI responsibly. The organizations gaining real advantage aren’t experimenting the most. They’re scaling with control, consistency, and accountability. Governance isn’t a constraint. It’s an enabler. Without it, teams move fast in different directions: tool sprawl, inconsistent data rules, uneven customer experiences, and shadow AI. The result isn’t just inefficiency: it’s reputational, operational, legal, and financial risk. Company-Wide Governance: The Guardrails Rules that apply everywhere, regardless of team or tool: → Data Rules Define what data can and cannot go into AI (PII, confidential info, regulated categories). “Use good judgment” isn’t policy. → Legal + Compliance Clarify what requires review, what claims/disclosures are allowed, and what standards apply to hiring/employee data. → Brand + Customer Promise Set standards for voice/tone, what commitments can be made, and prohibited language (guarantees, unauthorized discounts). → Security + Vendor Risk Maintain an approved tools list, require security review for new vendors, and enforce clear approval criteria. Department Governance: The Operating System Guardrails set boundaries. Departments define how work gets done inside them: → Approved Use Cases by Function Marketing, Sales, Support, HR: each needs specific guidance. Generic rules create gaps. → Human-in-the-Loop Checkpoints AI drafts; humans review for customer-facing, legal, or financial outputs. Define owners and escalation paths. → Prompt Standards + Templates Stop everyone winging it. Standardize proven prompts, document what works, and train teams for consistency. → Ownership + Accountability Every tool and use case needs an owner responsible for impact, quality, and decisions to expand or retire. Bottom line: Company governance without department execution becomes shelfware. Department execution without company governance creates risk and inconsistency. Build both, update regularly, and hold owners accountable, so you can move fast without breaking trust. Save this. It’s the difference between scaling AI and cleaning up after it.
211
69 Comments -
Cathy Hackl
Nokia • 176K followers
The next billion-dollar tech category isn't attention. It's intimacy. We are entering the Intimacy Economy, where emotional presence, trust, and connection are the new currency. I spoke with Global Banking & Finance Review about what I call Tech Intimacy: the intersection of AI, immersive technology, and human emotion. This is not a fringe trend. It is a fundamental shift in how we connect, communicate, and build relationships. Emotional connection is becoming the new growth engine and the smartest founders and investors are already building for it. Here is what is already unfolding: - A recent Harvard Business Review article titled How People Are Really Using Gen AI in 2025 reports that therapy and companionship is now the number one use case for generative AI, up from second place in 2024. It attributes this shift to the category nearly doubling its share of total AI usage, now accounting for 31% of all AI use cases, compared to 17% last year - Entire industries, from dating to wellness to entertainment, are shifting from performance to presence - Gen Z and Gen Alpha are redefining how intimacy, friendship, and trust work in digital spaces - Investors are backing a new wave of startups focused on emotional UX, AI connection, and synthetic relationships This is not about replacing human intimacy. It is about enhancing it and some of us are already building for the next layer of emotional infrastructure for the digital age. I am not a matchmaker. I am a futurist. And I believe the next wave of innovation will come from those who know how to design for emotional resonance, not just engagement metrics. Read the full article here: https://lnkd.in/enfMg4g7 #IntimacyEconomy #TechIntimacy #AI #FutureOfConnection #RelationshipTech #EmotionalTech #SpatialComputing #TechLoveLife #CathyHackl #FutureofDating #DatingApps #MetaAI #OpenAI
141
33 Comments -
Alice Goguen Hunsberger
Musubi • 14K followers
If you're looking to understand more about Responsible AI, Trust & Safety, and Public Interest Tech, then All Tech Is Human just released the guide for you. It's got great insights on the state of the industry, specific career paths to pursue (even job descriptions!) and interviews with leaders in the space (I'm honored to be one this time around). See the full report here: https://lnkd.in/eauKBH4j Huge congrats to ATIH for putting this together, I know it's a lot of work and I'm sure it will be really useful for lots of people.
112
1 Comment -
Lachie Tod
Order Editing • 10K followers
When your software sits at the center of enterprise operations, trust is non-negotiable. This week, we reached an important milestone at Order Editing. After more than three months of preparation across policy work, gap analysis, control testing, and auditor reviews, we are officially SOC 2 Type 1 compliant. I had the privilege of owning this project end to end. One of the realities of a fast-growing startup is wearing multiple hats. This time, it meant taking on a compliance initiative of real scale for the first time. It was demanding, unfamiliar, and genuinely rewarding. This effort was about far more than completing an audit. It required close coordination across engineering and account management, reshaping parts of our onboarding and internal processes, working with leadership and external auditors, and implementing Vanta as our compliance platform. In doing so, we did not just establish a SOC 2 framework. We materially improved how the business operates day to day. Clearer structure. Stronger documentation. Better access controls. More disciplined follow-through across the company. While I had ownership of the project, it was not a solo effort. The team stepped up across security reviews, infrastructure improvements, policy creation, evidence collection, and internal testing, all while continuing to support merchants through one of our busiest periods. What this reinforced for me is that SOC 2 is not just an audit milestone. It is a marker of operational maturity. You cannot reach this standard unless your systems, habits, and internal discipline already reflect what enterprise software demands. This milestone matters because Order Editing sits in the critical path of post-purchase operations. For enterprise brands, trust is non-negotiable. SOC 2 Type 1 signals that our approach to data handling and internal controls meets enterprise expectations. SOC 2 Type 2 is already underway. We now have a strong foundation to build from. #shopify #Soc2 #orderediting #growth #tech
76
16 Comments
Explore top content on LinkedIn
Find curated posts and insights for relevant topics all in one place.
View top content