Good morning. It's Tuesday, January 6, and we're covering NVIDIA’s CES releases, AI drug making, Maduro deepfakes, and more. First time reading? Join a community of 600k+ people keeping up with the future of tech. Sign up here.
🚀 NVIDIA Launches Powerful New Rubin Chip Architecture
NVIDIA has unveiled the Rubin platform, a rack-scale AI system built from six custom chips designed to function as a single AI supercomputer. Aimed at “AI factories,” Rubin targets large-model inference and training with up to 10× lower cost per token and significantly higher performance than its predecessor, Blackwell. The chips span compute, networking, and infrastructure.
🚗 NVIDIA Launches “Think Like a Human” Open-Source AV Models
NVIDIA introduced Alpamayo, a suite of open-source AI models and tools to advance autonomous vehicle development. At its core is Alpamayo 1, a 10B-parameter model that uses vision-language-action reasoning to interpret driving scenes and explain decisions—like a human driver might. The release includes simulation tools and datasets aimed at solving rare, complex edge cases.
👨💼 Meta AI Boss Sparks Talent Fears
Yann LeCun, former AI chief at Meta Platforms, warned the company risks losing researchers after appointing 29-year-old Alexandr Wang as chief AI officer. LeCun told the Financial Times Wang lacks experience running research teams, despite leading new AI unit TBD Labs. Wang, co-founder of Scale AI, joined Meta after it bought a 49% stake in his firm.
👩💻 Alexa Enters Chat Wars
Amazon has launched Alexa.com, bringing its Alexa+ AI assistant to web browsers for Early Access users. The site lets users get answers and complete tasks like shopping, planning, and smart home control. Alexa.com syncs context, preferences, and conversations across devices. Amazon says usage data shows higher engagement since Alexa+ launched nine months ago.
🪧 POWERED BY MULTIVERSE COMPUTING
Cut Compute Costs By 50% With CompactifAI
CompactifAI by Multiverse Computing compresses LLMs by up to 80% while maintaining accuracy, delivering up to 50% cost and energy savings for companies like IBM, Bank of Canada, and Moody’s. Run your models on fewer GPUs or even on the edge when you deploy CompactifAI to scale your AI workloads!
Try CompactifAI for Free and Realize Immediate Savings!
📺️ FROM THE LIVE SHOW
"Reflection is building the U.S. DeepSeek.”@MishaLaskin, CEO of @reflection_ai, on why the U.S. needs frontier-level open models.“We’re building frontier open models of Western provenance. It’s important to have small models people can tinker with, but it’s even more… pic.twitter.com/fVzGvdcENr
@ForwardFutureView post on X🦾 ROBOTICS
Boston Dynamics’ Atlas Gets Superhuman Agility As Factory Work Comes Into Focus
Boston Dynamics has unveiled major upgrades to its Atlas humanoid robot, showcasing fluid, acrobatic movement and expanded dexterity during a recent CBS News 60 Minutes segment. The latest version can cartwheel, pivot its torso without turning, and recover from the floor using only its feet—enabled by a redesigned body with continuous joint rotation and fewer failure-prone wires.
Powered by AI running on NVIDIA chips, Atlas is trained through teleoperation, where humans use virtual reality to teach tasks like stacking cups or tying knots. Company leaders say the goal is not to mimic human limits, but to build “superhuman” robots suited for industrial environments, while cautioning that widespread deployment will take time. → Read the full article here.
👨🔬 DRUG DISCOVERY
AI Is Rapidly Rewriting How Drugs Are Discovered
Artificial intelligence is moving from experimentation to core infrastructure in pharmaceutical research, dramatically accelerating how new drugs are designed, tested, and trialed. Major drugmakers and AI-native biotechs are using transformer models, digital simulations, and synthetic patients to cut years and billions of dollars from development timelines. The shift promises higher success rates in clinical trials—and could eventually reshape the balance of power between traditional pharma companies and AI labs.
AI is no longer a supporting tool in drug development—it is becoming the organising logic of the industry, compressing timelines and improving odds in a business long defined by risk and failure. If gains in trial efficiency and success rates hold, drugmaking could become less speculative and more scalable, unlocking sustained investment and faster access to new medicines. → Read the full article here. (Paywall)
🔲 SEMICONDUCTORS
Furiosa AI Begins Mass Production of First Inference Chip, RNGD
Furiosa AI, a South Korean AI chip startup, says it will begin mass production of its first integrated circuit, RNGD, in January 2026. Founded in 2017 by former AMD and Samsung engineer June Paik, the company is positioning its inference-focused accelerator as a lower-power alternative to Nvidia’s dominant GPUs.
RNGD is based on Furiosa’s proprietary Tensor Contraction Processor (TCP) architecture, which treats tensor contraction as a native operation rather than relying on fixed matrix multiplication. The move comes after Furiosa rejected an $800 million acquisition offer from Meta and raised $125 million in Series C funding last year. → Read the full article here.
🛰️ NEWS
What Else is Happening
✨ CES 2026’s Standout Tech Reveals: Early announcements spotlight foldable phones, AI-powered fridges, wireless ultra-thin TVs, and quirky gadgets, showing how everyday tech is getting smarter, thinner, and more personalized this year.
🛠️ Hyundai Eyes Robot-Built Cars: Hyundai is training humanoid robots from Boston Dynamics to work on car assembly lines in Georgia. The Atlas robot is being tested for tasks like assembly, parts sorting, and inspection.
🤥 AI Fakes Swirl After Maduro Capture: After reports of Maduro’s capture, altered photos and AI-generated images spread rapidly online. Some posts went viral hours before verified reporting confirmed details of his transfer to U.S. custody.
👥 SaaStr Replaces Sales With AI Agents: Founder Jason Lemkin says 20 autonomous agents now handle sales work once done by humans, betting scalable software beats costly, high-turnover sales hires.
🤖 Falcon H1R Challenges AI Scaling: Abu Dhabi’s TII unveiled a 7B open model that outperforms much larger rivals in reasoning, signaling a shift from brute-force size to smarter AI architectures.
🔮 Khosla Predicts AI Doctors: Billionaire Vinod Khosla says artificial intelligence will replace 80% of doctors’ tasks, handling diagnosis and decisions while humans focus on empathy and complex care.
📟 PROMPT OF THE WEEK
The "Pareto Principle" Year Planner
Act as an expert executive coach and productivity strategist. I am a [Insert Job Title] working in [Insert Industry]. I want to structure my upcoming work year using the Pareto Principle (80/20 rule) to maximize my impact and potential for a raise or promotion, while minimizing 'busy work.'
1. Analyze my role: Based on my job title, list the 3-5 high-leverage activities that likely drive the most value for my company.
2. Audit the low-value: List 3-5 common administrative or low-leverage tasks in this role that I should attempt to automate, delegate, or delete this year.
3. The Q1 Roadmap: Create a week-by-week focus plan for the first three months of the year. Focus on achieving one major 'Big Rock' project that aligns with the high-leverage activities you identified.🗒️ FEEDBACK
Help Us Get Better
What did you think of today's newsletter? |
That’s a Wrap!
🔈️ Want to advertise with Forward Future? Reach 600K+ AI enthusiasts, tech leaders, and decision-makers. Let’s talk—just reply to this email.
🧑💻 The AI-Fluent Job Board: Cut through the noise of generic job boards and hire AI-fluent talent, fast → Learn more.
🛰️ Want more Forward Future? Follow us on X for quick updates, subscribe to our YouTube for deep dives, or add us to your RSS Feed for seamless reading.
Thanks for reading today’s newsletter—see you next time!
Matthew Berman & The Forward Future Team
🧑🚀 🧑🚀 🧑🚀 🧑🚀 🧑🚀 🧑🚀
