In this article, you will learn how to distinguish agentic workflows from autonomous agents by focusing on who owns control flow — a human writing…
Agentic Workflow vs. Autonomous Agent: What’s the Difference? 投稿を読む »
Artificial intelligence is rapidly reshaping retail, but not in the ways consumers might immediately notice. The biggest transformation may not be flashy virtual try-ons or chatbot shopping assistants, but in how decisions are made behind the scenes: how products surface in search results, how inventory moves through supply chains, how engineers ship code faster, and how retailers respond to customer behavior in real time. As legacy retailers navigate a fragmented and hyper-competitive landscape, AI is becoming an operating philosophy. At Macy’s, that philosophy is more often defined by what senior director of engineering Murali Murugan describes as an “AI-first” approach. “AI first isn’t about adding intelligence on top,” Murugan says. “It’s about redesigning how decisions happen so the business moves faster and every experience feels more relevant by default.” Rather than layering AI onto existing workflows, Macy’s is embedding intelligence directly into systems that include personalization, search, operational planning, and software development itself. The company’s strategy is reflective of a larger shift taking place across retail: moving from isolated AI pilots toward integrated systems designed to compress, as Murugan puts it, “the gap between the signal and the action.” Early efforts focused on narrow, high-impact use cases like search recommendations and customer engagement, where measurable gains in conversion and reduced friction quickly built internal momentum. “Once we established the quick wins, scaling was a business decision, not a technology debate anymore,” he says. That momentum is now extending into conversational commerce through tools like Ask Macy’s, an AI-powered shopping assistant designed to act more like a personal stylist than a traditional search bar. Whether for a prom, a vacation, or a last-minute event, customers can describe what they need conversationally and receive curated recommendations informed by past purchases, preferences, and context. Still, the company sees AI as more of an invisible layer augmenting human judgment than a replacement for it. The long-term vision is retail that feels increasingly seamless, adaptive, and personalized, powered by systems customers may never even notice are there. “The real transformation in this all comes from continuous improvement,” Murugan says. “It’s about learning from the mistakes, quickly adapting to the newer technology standards that are coming into play, timing, and execution which compound into a meaningfully better customer experience.” This webcast is produced in partnership with Infosys. This content was produced by Insights, the custom content arm of MIT Technology Review. It was not written by MIT Technology Review’s editorial staff. It was researched, designed, and written by human writers, editors, analysts, and illustrators. This includes the writing of surveys and collection of data for surveys. AI tools that may have been used were limited to secondary production processes that passed thorough human review.
Repositioning retail for the AI era 投稿を読む »
This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. Europe’s extreme heat is shutting down power plants Europe is in the middle of a record-breaking heat wave, and the grid is being pushed to its limits as people turn to fans and air-conditioning to try to stay cool. But some power plants won’t be online to help handle the load. The main source of stress is increased demand, largely driven by cooling. And the challenges are only expected to worsen as climate change brings more frequent and intense heat waves. Find out how rising temperatures are stretching power supplies—and how utilities can adapt. —Casey Crownhart What Europe’s heat wave means for the power grid Grid planning in the age of climate change generally means that we need a lot more supply, and quickly. But one interesting facet to this challenge is that in some places, seasonal patterns are shifting, compounding the difficulty of meeting demand. Europe has historically seen its grid peak in the winter when electric heating is widespread. So some planned outages happen in the spring and into the summer, which is affecting the supply right now. But a growing need for air-conditioning will alter the balance. Read the full story on how climate change is reshaping electricity demand. —Casey Crownhart This story is from The Spark, our weekly newsletter giving you the inside track on all things climate. Sign up to receive it in your inbox every Wednesday. IBM unveils chip technology that could help extend Moore’s Law another decade IBM has built a new prototype chip with around 100 billion transistors on an area the size of a fingernail. That’s twice the density of the company’s previous state-of-the-art technology announced in 2021. And the design could pave the way for faster and more energy-efficient computers for years to come. In the last fifteen years, transistors have been shrunk close to their limits. They can’t get smaller without their function deteriorating. IBM’s new chip resolves this with an approach familiar to urban planners: building up. Here’s how the strategy is bringing new hope to the technology industry. —Sophia Chen The must-reads I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology. 1 Anthropic says Alibaba “illicitly” extracted Claude’s capabilities It claims the Chinese firm ran a “brazen” campaign to access the model. (BBC)+ It says it’s the “largest known distillation attack” on the company. (CNBC)+ The technique trains a weaker model on a stronger one’s outputs. (FT $)+ Anthropic previously accused other Chinese rivals of using it. (CNN)+ But it’s still feuding with the White House. (MIT Technology Review) 2 NASA has detected possible chemical signatures of ancient life on MarsThe Perseverance rover spotted complex carbon on rocks. (New Scientist $)+ The molecules are typically associated with dead organisms. (Guardian)+ The US has lost its lead in the hunt for alien life. (MIT Technology Review) 3 The EU has joined a US pact to stop relying on Chinese AIMuch of the rest of the world seems to still be a battleground for control. (FT $)+ China is expanding its AI push in the Global South to counter the US. (The Wire China)+ Chinese AI experts are freaking out about the AI arms race. (Wired $) 4 OpenAI and Broadcom have unveiled their first jointly designed AI chipJalapeño is built to power large-scale AI systems like ChatGPT. (NYT $)+ It’s part of OpenAI’s push to “build the full stack.” (CNBC) 5 A new report shows ICE has built a vast hi-tech surveillance systemIt includes facial recognition, drones, and data scraping.(Guardian)+ Is the Pentagon allowed to surveil citizens with AI? (MIT Technology Review) 6 Electronics can now be printed onto living tissueWhich could enable smart implants and ingestible diagnostics. (The Economist $) 7 The data center boom is sparking a third wave of inflation Demand for memory chips is pushing prices higher.(WSJ $) 8 Companies are scrambling to curb spending on AI token “chewing”Accenture data shows non-technical staff are draining budgets. (404 Media) 9 Claude Design is creating a bland wave of website uniformityThe AI tool is homogenizing the internet’s aesthetic. (The New Yorker $) 10 Elon Musk has lost his trillionaire statusThanks to SpaceX stock coming back to Earth. (Business Insider) Quote of the day “Tom Brown is not being a weirdo like Dario and can actually engage.” —A person directly familiar with calls between the Trump administration and Anthropic tells Wired that they’ve improved since cofounder Tom Brown replaced CEO Dario Amodei in the talks. One More Thing TONY LUONG The quest to learn if our brain’s mutations affect mental health For years, scientists searching for the roots of conditions like schizophrenia, autism, and Alzheimer’s have focused on single genes. But the real source may lie in a more complex genetic puzzle inside the brain. Mike McConnell has spent decades exploring the idea that neurons do not all share identical DNA, and that these differences could help explain psychiatric disease. His work has contributed to evidence that brain cells can form a “genetic mosaic,” with mutations that vary across the brain. Discover how this could reshape our understanding of mental illness. —Roxanne Khamsi We can still have nice things A place for comfort, fun, and distraction to brighten up your day. (Got any ideas? Drop me a line.) + This classical reimagining of the Super Mario soundtrack is exquisite.+ At long last, we can calculate the fuel efficiency of launching our enemies into the Sun.+ Before CGI, explosions were an art form. This compilation of classic practical effects is pure action-movie nostalgia.+ Cambridge botanists lovingly recreated a 336-year-old garden to honor the “father of natural history.” (Big thanks to reader Peter Ryan for the find!)
The Download: Europe’s heat wave hits the grid, and IBM’s chip targets Moore’s Law 投稿を読む »
DeepReinforce has released Ornith-1.0, an open-source model family built for agentic coding. The lineup spans four sizes, from a 9B dense model to a 397B mixture-of-experts flagship. Every checkpoint ships under the MIT license on Hugging Face. The models are post-trained on top of pretrained Gemma 4 and Qwen 3.5. Most coding agents pair a model with a fixed, human-designed harness. Ornith-1.0 instead learns to write its own. The DeepReinforce research team reports state-of-the-art results among open models of comparable size. TL;DR Ornith-1.0 ships in 9B, 31B, 35B-MoE, and 397B-MoE sizes under MIT, built on Gemma 4 and Qwen 3.5. The model learns its own scaffold during RL, jointly optimizing the harness and the solution. Ornith-1.0-397B tops Claude Opus 4.7 on both headline benchmarks, but not Opus 4.8 or the larger GLM-5.2-744B. Three layers — fixed trust boundary, deterministic monitor, frozen LLM judge — guard against reward hacking. What is Ornith-1.0? Ornith-1.0 is a set of reasoning models tuned for coding agents. The variants are 9B Dense, 31B Dense, 35B MoE, and 397B MoE. The 35B model is mixture-of-experts and activates roughly 3B parameters per token. FP8 and GGUF builds are also published for faster local serving. Each model is a reasoning model. Replies open with a <think> block before the final answer. The serving recipes enable a reasoning parser, so that trace returns in a separate reasoning_content field. The models also emit well-formed tool calls for agent loops. Deployment is straightforward. The 9B model is about 19GB in bf16 and serves on a single 80GB GPU. Serving recipes target vLLM, SGLang, and Transformers. Each model exposes an OpenAI-compatible endpoint. Standard agent frameworks therefore work without code changes. Interactive Explainer </button> <button class=”btn gho” id=”resetBtn”>Reset</button> </div> <div class=”stepout” id=”stepOut”>Step 0 — untrained policy with a fixed, hand-written harness.</div> </div> <!– PANEL 2: BENCH –> <div class=”panel” data-panel=”bench”> <div class=”lead”>Vendor-reported scores from DeepReinforce. Pick a model tier and a benchmark. Ornith is highlighted in green. Higher is better.</div> <div class=”seg”><span class=”lab”>Model tier</span> <div class=”chip on” data-tier=”t397″>397B flagship</div> <div class=”chip” data-tier=”t35″>35B MoE</div> <div class=”chip” data-tier=”t9″>9B dense</div> </div> <div class=”seg” id=”benchChips”><span class=”lab”>Benchmark</span></div> <div class=”chart” id=”chart”></div> <div class=”foot-note” id=”benchNote”></div> </div> <!– PANEL 3: DEFENSES –> <div class=”panel” data-panel=”def”> <div class=”lead”>A model that writes its own scaffold could cheat the verifier. DeepReinforce describes three defense layers. Tap each to expand.</div> <div class=”layers”> <div class=”layer open”><div class=”lh”><span class=”num”>1</span><span class=”lt”>Fixed trust boundary</span><span class=”more”>tap</span></div><div class=”lb”>The environment, tool surface, and test isolation are immutable and outside the model’s reach. The model evolves only its inner policy scaffold — memory, error-handling, and orchestration logic.</div></div> <div class=”layer”><div class=”lh”><span class=”num”>2</span><span class=”lt”>Deterministic monitor</span><span class=”more”>tap</span></div><div class=”lb”>A rule-based monitor flags any attempt to read withheld paths, modify verification scripts, or invoke unsanctioned tools. Such trajectories get zero reward and are excluded from the advantage computation.</div></div> <div class=”layer”><div class=”lh”><span class=”num”>3</span><span class=”lt”>Frozen LLM judge</span><span class=”more”>tap</span></div><div class=”lb”>Because intent-level gaming can happen inside the allowed tool surface, a frozen LLM judge acts as a veto on top of the verifier — not as the primary reward signal.</div></div> </div> </div> <div class=”ftr”><span>Source: <a href=”https://deep-reinforce.com/ornith_1_0.html” target=”_blank” rel=”noopener”>deep-reinforce.com</a> · MIT licensed · numbers vendor-reported</span><span><b>Marktechpost</b> · AI Dev Signals</span></div> <script> (function(){ var root=document.getElementById(‘mtp-ornith-demo’); /* tabs */ root.querySelectorAll(‘.tab’).forEach(function(t){ t.addEventListener(‘click’,function(){ root.querySelectorAll(‘.tab’).forEach(function(x){x.classList.remove(‘on’)}); root.querySelectorAll(‘.panel’).forEach(function(x){x.classList.remove(‘on’)}); t.classList.add(‘on’); root.querySelector(‘.panel[data-panel=”‘+t.dataset.p+’”]’).classList.add(‘on’); resize(); }); }); /* loop sim */ var step=0,reward=0.08,timer=null; var scaffs=[ ‘Baseline harness: linear retries, no memory.’, ‘Adds scratchpad memory across tool calls.’, ‘Adds error-triage branch before re-edit.’, ‘Reorders: read tests, then plan, then patch.’, ‘Caches sub-results; prunes dead branches.’, ‘Task-specific orchestration emerges automatically.’]; var outs=[ ‘Fixed harness, no learning yet.’, ‘Fewer redundant file reads observed.’, ‘Recovers from failed edits more often.’, ‘Higher first-pass test success.’, ‘Shorter trajectories, same accuracy.’, ‘Stable high-reward scaffold selected.’]; var nodes=root.querySelectorAll(‘.node’); function lightSeq(cb){ var i=0;nodes.forEach(function(n){n.classList.remove(‘act’)}); var iv=setInterval(function(){ nodes.forEach(function(n){n.classList.remove(‘act’)}); nodes[i].classList.add(‘act’);i++; if(i>=nodes.length){clearInterval(iv);setTimeout(function(){nodes.forEach(function(n){n.classList.remove(‘act’)});cb&&cb();},260);} },220); } function doStep(){ if(step>=5){return;} step++; lightSeq(function(){ reward=[0.08,0.27,0.43,0.58,0.69,0.77][step]; root.querySelector(‘#rFill’).style.width=(reward*100)+’%’; root.querySelector(‘#rVal’).textContent=reward.toFixed(2); root.querySelector(‘#scaffTxt’).textContent=scaffs[step]; root.querySelector(‘#outTxt’).textContent=outs[step]; root.querySelector(‘#stepOut’).innerHTML=’Step ‘+step+’ — <b>scaffold mutated</b>; reward propagated to both stages.’; resize(); }); } root.querySelector(‘#stepBtn’).addEventListener(‘click’,doStep); root.querySelector(‘#autoBtn’).addEventListener(‘click’,function(){ if(timer){clearInterval(timer);timer=null;this.textContent=’Auto-run ‘;return;} this.textContent=’Pause ‘;var b=this; timer=setInterval(function(){if(step>=5){clearInterval(timer);timer=null;b.textContent=’Auto-run ‘;}else{doStep();}},1400); }); root.querySelector(‘#resetBtn’).addEventListener(‘click’,function(){ if(timer){clearInterval(timer);timer=null;root.querySelector(‘#autoBtn’).textContent=’Auto-run ‘;} step=0;reward=0.08; root.querySelector(‘#rFill’).style.width=’8%’; root.querySelector(‘#rVal’).textContent=’0.08′; root.querySelector(‘#scaffTxt’).textContent=scaffs[0]; root.querySelector(‘#outTxt’).textContent=’Press “Run training step” to begin.’; root.querySelector(‘#stepOut’).innerHTML=’Step 0 — untrained policy with a fixed, hand-written harness.’; resize(); }); /* benchmark data (vendor-reported) */ var BENCHES=[‘Terminal-Bench 2.1′,’SWE-Bench Verified’,’SWE-Bench Pro’,’SWE-Bench Multilingual’,’NL2Repo’,’ClawEval Avg’]; var DATA={ t397:{label:’Ornith-1.0-397B’,hero:’Ornith-1.0-397B’, models:[‘Ornith-1.0-397B’,’Qwen3.5-397B’,’Qwen3.7-Max’,’GLM-5.2-744B’,’Minimax-M3-428B’,’DeepSeek-V4-Pro-1.6T’,’Claude Opus 4.7′,’Claude Opus 4.8′], vals:[[77.5,53.5,73.5,81.0,64,64,70.3,85],[82.4,76.4,80.4,null,null,80.6,80.8,87.6],[62.2,51.6,60.6,62.1,59,55.4,64.3,69.2],[78.9,69.3,78.3,null,null,76.2,null,null],[48.2,36.8,47.2,48.9,42.1,null,null,69.7],[77.1,70.7,65.2,null,null,75.8,78.2,null]]}, t35:{label:’Ornith-1.0-35B-A3B’,hero:’Ornith-1.0-35B-A3B’, models:[‘Ornith-1.0-35B-A3B’,’Qwen3.5-35B-A3B’,’Qwen3.6-35B-A3B’,’Gemma4-31B’,’Qwen3.5-397B’], vals:[[64.2,41.4,52.5,42.1,53.5],[75.6,70,73.4,52,76.4],[50.4,44.6,49.5,35.7,51.6],[69.3,60.3,67.2,51.7,69.3],[34.6,20.5,29.4,15.5,36.8],[69.8,65.4,68.7,48.5,70.7]]}, t9:{label:’Ornith-1.0-9B’,hero:’Ornith-1.0-9B’, models:[‘Ornith-1.0-9B’,’Qwen3.5-9B’,’Qwen3.5-35B-A3B’,’Gemma4-12B’,’Gemma4-31B’], vals:[[43.1,21.3,41.4,21,42.1],[69.4,53.2,70,44.2,52],[42.9,31.3,44.6,27.6,35.7],[52,39.7,60.3,32.5,51.7],[27.2,16.2,20.5,10.3,15.5],[63.1,53.2,65.4,32.5,48.5]]} }; var curTier=’t397′,curB=0; var bchips=root.querySelector(‘#benchChips’); BENCHES.forEach(function(b,i){ var c=document.createElement(‘div’);c.className=’chip’+(i===0?’ on’:”);c.textContent=b;c.dataset.b=i; c.addEventListener(‘click’,function(){curB=i;bchips.querySelectorAll(‘.chip’).forEach(function(x){x.classList.remove(‘on’)});c.classList.add(‘on’);draw();}); bchips.appendChild(c); }); root.querySelectorAll(‘.chip[data-tier]’).forEach(function(c){ c.addEventListener(‘click’,function(){curTier=c.dataset.tier;root.querySelectorAll(‘.chip[data-tier]’).forEach(function(x){x.classList.remove(‘on’)});c.classList.add(‘on’);draw();}); }); function draw(){ var d=DATA[curTier];var row=d.vals[curB];var chart=root.querySelector(‘#chart’);chart.innerHTML=”; var max=Math.max.apply(null,row.filter(function(v){return v!=null})); d.models.forEach(function(m,i){ var v=row[i];var hero=(m===d.hero); var div=document.createElement(‘div’);div.className=’row’+(hero?’ hero’:”)+(v==null?’ na’:”); div.innerHTML='<div class=”nm”>’+m+'</div><div class=”bt”><div class=”bf”></div></div><div class=”vl”>’+(v==null?’n/a’:v)+'</div>’; chart.appendChild(div); (function(bf,val){setTimeout(function(){bf.style.width=(val==null?0:(val/max*100))+’%’;},40);})(div.querySelector(‘.bf’),v); }); root.querySelector(‘#benchNote’).textContent=’Benchmark: ‘+BENCHES[curB]+’. Bars scaled to the highest score shown. “n/a” = not reported by the vendor. Self-reported, not independently verified.’; resize(); } draw(); /* defenses accordion */ root.querySelectorAll(‘.layer’).forEach(function(l){ l.addEventListener(‘click’,function(){l.classList.toggle(‘open’);resize();}); }); /* auto-resize for WordPress iframe */ function resize(){ try{ var h=root.offsetHeight+40; if(window.parent){window.parent.postMessage({type:’mtp-ornith-height’,height:h},’*’);} }catch(e){} } window.addEventListener(‘load’,resize); setTimeout(resize,300); window.addEventListener(‘resize’,resize); })(); </script> </div> ” style=”width:100%;border:0;display:block;min-height:600px;overflow:hidden” height=”600″ scrolling=”no” loading=”lazy” title=”Ornith-1.0 Interactive Explainer”> The Self-Scaffolding Idea Most coding agents rely on a scaffold, also called a harness. A scaffold wraps the model with memory, tools, error handling, and orchestration logic. AI teams usually hand-design one scaffold per task category. Ornith-1.0 treats the scaffold as a learnable object instead. During reinforcement learning, the scaffold co-evolves with the model’s policy. Each RL step runs in two stages. First, the model reads the task and its previous scaffold. It then proposes a refined scaffold. Second, it uses that scaffold and the task to generate a solution rollout. Reward from the rollout flows back to both stages. So the model is optimized to author orchestration, not just answers. Over training, higher-reward scaffolds are mutated and selected automatically. Per-task strategies emerge without hand-engineered harness design. Training also runs asynchronously, using a pipeline-RL setup. A staleness weight downweights older, off-policy tokens and drops them past a threshold. The optimization uses a token-level GRPO objective. Guarding Against Reward Hacking Letting a model write its own scaffold invites reward hacking. A scaffold could read visible test files and hardcode expected outputs. It could also copy an oracle solution sitting in the environment. DeepReinforce team describes three defense layers. The outer trust
Europe is in the middle of a record-breaking heat wave, and the grid is being pushed to its limits as people turn to fans and air-conditioning to try to stay cool. Some power plants won’t be online to help handle the load. On June 23, France saw its hottest day since record-keeping began in 1947. Temperatures climbed to over 44 °C (111 °F), and overnight temperatures remained unusually high. This prolonged hot weather warmed up the water in some rivers across the country, a problem for the many nuclear plants that rely on those bodies of water for cooling. One reactor has already shut down, and others are being ramped down or will see limitations later in the week. Unit two at the Golfech nuclear power plant in southern France shut down at about 11:45 p.m. on June 22 when the river used to cool the plant got too hot. The move was a precautionary measure, according to Brid Nelligan, a spokesperson for EDF, the plant’s owner and operator. The power plant takes in water from the Garonne River and then returns most of it to the river at slightly higher temperatures after using it to cool equipment. French regulations limit the temperature of that return stream, so the warm water (it was expected to reach 28 °C, or around 82 °F) forced the operator to shut down the plant. EDF, which operates France’s entire nuclear fleet, is also limiting the output of other reactors across the country—one reactor at the Nogent-sur-Seine power plant was ramped down as of Tuesday, and more will follow later in the week, Nelligan says. Extreme heat has affected France’s nuclear industry before. At least seven gigawatts’ worth of nuclear energy was forced to shut down across the country during a heat wave in July 2025, according to data from Ember Energy. That’s more than the entire grid of Ireland. This time, power plant outages and limitations aren’t expected to be drastic enough to affect the ability to meet demand in France, according to RTE, operator of the national electric grid. Nuclear power has made most of the headlines during this heat wave, but other forms of electricity generation face similar challenges. Hydropower plants frequently run into problems when dry conditions lower the amount of water available to generate energy and force them to decrease or shut off operations. In the first five months of 2025, high temperatures and low water conditions cut hydropower supplies in Europe by 13% compared with the year before. Even established coal and natural-gas plants can be challenged by high temperatures. Hot weather can stress equipment and limit the efficiency of cooling towers. Five gas plants across the UK have reported output reductions due to the conditions, cutting a total of about 2.5 gigawatts from the power supply. Increased demand, largely driven by cooling, is the main factor stressing Europe’s power grid, says Jean-Paul Harreman, director of Montel, an energy intelligence provider, via email. Even countries that haven’t historically relied much on cooling technologies are turning to them now—the number of UK homes that use air-conditioning has roughly doubled since 2022. Around the world, the challenges heat presents for the grid are only expected to get worse as climate change brings more frequent and intense heat waves. Globally, energy use for cooling is set to double by 2050 relative to 2023 levels, according to the International Energy Agency. “Utilities can adapt by planning for summer peaks, making cooling demand more flexible, reinforcing grids for high temperatures, deploying batteries and demand response, and climate-proofing power plants’ cooling systems,” says Simone Tagliapietra, senior fellow at Bruegel, an economic and policy think tank, via email. But those changes could be expensive. Earlier this year, EDF shared a climate-change vulnerability assessment for its business, including nuclear and hydropower operations across France. Upgrades are expected to cost about €600 million per year (about $680 million) over the next 15 years. Meanwhile, high temperatures are expected to continue across much of Europe through the end of the week.
Europe’s extreme heat is shutting down power plants 投稿を読む »
This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. Introducing: the Engineering issue We can’t fix everything, but we can be ambitious. We can take on the challenge of making the world better through human ingenuity. That’s what the new Engineering issue of MIT Technology Review is all about. Sometimes the challenges we face are giant, like tunneling beneath the seafloor. Some exist at the nanoscale, as with a new ASML machine powering the future of chipmaking. Others represent problems at a planetary scale and in truly unknown territory, like replicating a volcano’s mechanism to cool the Earth on purpose. These incredible engineering stories show we can come together to get to work and, when the smoke clears, find we’ve made real progress. Subscribe now to read all of them—and more—in the full print issue. Stripe, Anthropic, and OpenAI are backing an effort to stop respiratory infections The common cold comes for us all—often more than once a year. And there is no way to prevent it. The best you can do is take vitamin C and stay away from people with the sniffles. Now, the payment company Stripe is funding a new $500-million nonprofit aiming to prevent both the common cold and the flu. Its eventual goal is to get rid of respiratory viruses altogether. Anthropic, OpenAI, and Bill Gates have also backed the venture, which will investigate whether modern technologies can counter the common cold and the flu. Dive into the nonprofit’s plans. —Antonio Regalado MIT Technology Review Narrated: inside the hunt for the most dangerous asteroid ever As asteroid 2024 YR4 hurtled toward Earth, astronomers determined that this massive rock posed a higher risk of impact than any object of its size in recorded history. Then, just as quickly as history was made, experts declared that the danger had passed. This is the inside story of the network of global scientists who found, followed, planned for, and finally dismissed the most dangerous asteroid ever discovered —all under the tightest of timelines and with the highest of stakes. —Robin George Andrews This is our latest story to be turned into an MIT Technology Review Narrated podcast, which we publish each week on Spotify and Apple Podcasts. Just navigate to MIT Technology Review Narrated on either platform, and follow us to get all our new content as it’s released. The must-reads I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology. 1 China has taken the US’s crown for the world’s fastest supercomputer Shenzhen’s LineShine overtook California’s El Capitan. (Axios)+ China had not had a machine at the top of the list since 2017. (NYT $)+ But the supercomputer race isn’t geared for AI work. (Reuters $) 2 Mythos reportedly found flaws in classified US government systemsA US official said Anthropic’s model identified certain vulnerabilities. (AP News)+ The model has now been suspended over US security concerns. (BBC)+ The NSA has lost access to Anthropic’s tools in fallout. (Engadget)+ The feud raises new questions about AI safety. (MIT Technology Review) 3 A US pilot reported seeing Iranian drones swarm in “jellyfish” formationWhich would represent an alarming advance in Iranian drone capabilities. (CNN)+ The US is heading toward a drone-filled future. (MIT Technology Review) 4 Mark Zuckerberg directed Meta to create a prediction markets appIt will be similar to Polymarket and Kalshi. (NYT $)+ But won’t let users wager real money. (The Verge) + Another new app, Meta Photos, will create media with AI. (Reuters $) 5 SpaceX’s “Starfall” just launched a secretive test flightThe orbital delivery spacecraft blasted off for the first time yesterday. (Axios)+ It could also support space manufacturing. (New Scientist $) 6 Alibaba has sued the US for being linked to the Chinese militaryIt wants to be removed from a Pentagon blacklist. (Reuters $) 7 Nvidia’s banned AI chips have doubled in price on China’s black marketThe DGX B300 now costs more than $1.1 million. (Financial Times $) 8 Tesla claims a driver “manually overrode self-driving” in a deadly crashIt said the accelerator was pressed “all the way to 100%.” (The Verge $) 9 The US science retreat has created an opportunity for EuropeBut questions about funding and innovation remain. (Nature)+ Trump has dealt many blows to US science. (MIT Technology Review) 10 Meta’s new smart glasses ditch Ray-Bans for Kylie Jenner Meta logos and Jenner designs have replaced the Ray-Ban branding. (Wired $) Quote of the day “It’s blasphemy against AI if you say it’s a bubble.” —SoftBank founder and CEO Masayoshi Son tells shareholders that the AI boom is still in its early stages, Reuters reports. One More Thing ERIK CARTER Video games are dividing South Korea They say StarCraft was the game that changed everything. When the science fiction strategy game arrived in South Korea in 1998, it wasn’t just a hit—it was an awakening. Out of 11 million copies sold worldwide, 4.5 million were in the country. The game was so popular that it triggered another boom: “PC bangs,” pay-as-you-go gaming cafés. StarCraft and PC bangs spoke to a generation of young South Koreans boxed in by economic anxiety and rising academic pressures. But they also sparked arguments about game addiction. They’ve led to feuds between government departments—and a national debate over policy. Read the full story. —Max S. Kim We can still have nice things A place for comfort, fun, and distraction to brighten up your day. (Got any ideas? Drop me a line.) + This archive lovingly documents the beautiful design of over 1,700 obsolete objects.+ Classic TV theme tunes like Hey Arnold! Have been revived in a musician’s marvellous samples.+ Marvel at the mind-boggling geometry of nature and see how bees perfectly construct honeycombs.+ Hear the ominous, deeply atmospheric tones of a custom string instrument built inside a plastic drainage pipe.
The Download: introducing the Engineering issue 投稿を読む »
In this article, you will learn why a large context window is not the same thing as agent memory, and how techniques like retrieval, compression,…
Context Windows Are Not Memory: What AI Agent Developers Need to Understand 投稿を読む »
The common cold comes for us all—often more than once a year. And there is no way to prevent it. The best you can do is take vitamin C and stay away from people with the sniffles. Now the payment company Stripe, founded by brothers Patrick and John Collison, says it will fund a new $500 million nonprofit whose goal is preventing both the common cold and the flu. Its eventual aim is to get rid of respiratory viruses altogether. The new organization, called Intercept, will use grants and investments to back prevention approaches, including vaccines, as well as large-scale air-cleaning systems for schools, offices, and other public spaces. In addition to Stripe, other funders include Anthropic, Flu Lab, and the OpenAI Foundation, as well as Bill Gates and several traders at the quantitative investing fund Jane Street Capital, according to an Intercept spokesperson. “I think we treat respiratory infections as a minor nuisance, but have really underweighted the burden that they impose on society,” says Nan Ransohoff, the Stripe executive leading the initiative along with Charlie Petty, a venture capitalist who joined Stripe this year. On average, people spend 5% of their lifetime fighting a cold or the flu, according to Ransohoff. Despite that, drug companies put relatively little effort into preventing colds. Part of the problem is that the sniffles are caused by more than 200 different viruses, according to the American Lung Association, with rhinoviruses being the most common culprits. There are so many that it typically doesn’t pay to try to stop any one of them with a vaccine. “When pharma companies look at it, it’s not as attractive as other things they could work on,” says Ransohoff. “So it hasn’t attracted the resources.” Stripe previously organized a $1.8 billion program called Frontier to encourage the development of carbon removal technology, as a way of countering climate change. Ransohoff says removing carbon from the atmosphere and getting rid of respiratory viruses are similar in that each is “technically possible” but they “lack commercial incentives.” The concept for Intercept took shape after Ransohoff started talking to David Veesler, a structural biologist and vaccine designer at the University of Washington, who argued that it’s possible to come up with broad countermeasures that work against many viruses at once. “He effectively sort of nerd-sniped me,” Ransohoff says of Veesler. “He convinced me that this is technically possible. He also helped me understand that some of the reasons that this hasn’t been done before was sort of an incentive problem.” Veesler says the growing tool kit available to scientists includes RNA drugs, antibodies, and computational protein design. For instance, one idea is to engineer virus-grabbing proteins that people could spray in their nasal passages, to catch viruses before they cause infection. “Most people just accept these viruses as a fact of life, and that got us thinking: Do we have to accept it?” says Veesler. “The more we thought about it, the more we realized that many of these problems have not been worked on with modern technologies.” The project takes inspiration from efforts to fight the covid-19 virus, where Veesler’s group was among those involved in the speedy development of vaccines, antiviral drugs, and antibodies. According to Ransohoff, Intercept’s advisors will include Peter Marks, a former top FDA official, as well as Moncef Slaoui, the pharmaceutical executive who led the US coronavirus vaccine effort, Operation Warp Speed. A key challenge for Intercept will be coming up with ways to counter many viruses at one time. That accounts for the interest in air-cleaning technology, such as using strong ultraviolet light to inactivate viruses. The idea, the group says, is to remove them from the air in the same way municipalities remove impurities from the water supply before it’s piped to people’s homes. The US funds about $6.5 billion a year in virus research through the National Institute of Allergy and Infectious Disease, or NIAID. But that agency’s budget hasn’t grown in recent years, leaving more room for private philanthropy. And Stripe’s Collison brothers have become some of the most reliable philanthropists in viral research. After giving away “fast grants” to help labs during the covid-19 pandemic, they later joined other donors who committed $650 million to establish the Arc Institute in Palo Alto, California, which has developed AI models for biological research. “The diversity of viruses is just too large and seems daunting, so people don’t even try,” says Veesler. “I’m happy that someone is ready to help scientists, not accepting the status quo, and doing something different.”
Stripe, Anthropic, and OpenAI are backing an effort to stop respiratory infections 投稿を読む »
AI is booming. New use cases are emerging each day. To capitalize on the technology’s potential, enterprises require data at scale. In many cases, though, the relevant information is blocked or unstructured, which limits its use by AI models. To understand this challenge, consider the foundation of the web itself. The web was not designed for the automated discovery and retrieval that new AI applications demand. Overcoming this inherent design constraint requires infrastructure. The next frontier in AI may depend on a new web data infrastructure layer that can enable models to discover and map this ever-expanding digital realm. This layer must be able to navigate hundreds of millions of existing web domains and billions of new URLs created each week, delivering real-time information and overcoming technical barriers. “The data suggests there’s far more data out there,” says Or Lenchner, CEO of Bright Data, a web data collection platform. “Think of the universe: It’s out there, but you don’t know what you don’t know.” Enabling access to fresh, relevant, and trustworthy data While early AI breakthroughs were driven by scaling training data and model size, organizations are now encountering a fundamental bottleneck: They need to keep pace with the dynamic, unstructured, and constantly evolving nature of web data in order to ground outputs in current and verifiable information. AI performance increasingly depends not just on model architecture but on a system’s compute, networking, retrieval, and data engineering capabilities—that is, the system’s ability to quickly and reliably retrieve data that is fresh, relevant, and trustworthy. Traditional model training relies on snapshots of information collected at a particular point in time. Training AI on such static data is no longer sufficient. To track fluctuations such as competitor pricing, consumer sentiment, and market trends, companies need a constant feed of new information, pulling data in real time along with relevant context. Their infrastructure must therefore be able to handle millions of simultaneous interactions across websites that vary by geography, language, format, and access rules. “If it can’t retrieve real-time information, it lacks context,” Lenchner says. “In a business setting, that’s not acceptable anymore. Stale answers lead to bad decisions and disappointed consumers.” Speed is not merely a matter of convenience; it’s a matter of necessity. Today’s organizations operate in environments where prices, inventory, markets, security threats, and customer behavior change continuously. Delayed data retrieval can reduce the usefulness of an otherwise sophisticated model. Using live, high-quality web data can also reduce AI hallucinations because the model has a more relevant knowledge base. This builds user trust. In fact, one survey found that 56% of AI practitioners said businesses need access to real-time web data to improve trust in AI outputs. To ensure the model runs efficiently and effectively, the information must also be pared down to the appropriate essentials. Despite the introduction of retrieval-augmented generation (RAG), where models pull in external data at the moment of a query, many AI systems still struggle to deliver outputs that are current, contextually relevant, and trustworthy in operational settings. According to Gartner, 60% of AI projects that are not supported by AI-ready data—accurate, structured, organized, and contextualized—will be abandoned by the end of the year. This is because large-scale retrieval alone does not solve the problem. As Lenchner puts it, “You need to retrieve data at scale, but also in real time. Latency becomes an issue because of the end user who is waiting for the output.” Accessing fresh, AI-ready data at scale introduces technical and structural challenges. In practice, many enterprise systems combine public web retrieval with APIs, licensed datasets, and proprietary internal data in their AI applications. Integrating these fragmented sources into a timely and usable knowledge layer requires specialized capabilities. Some research has found that 97% of AI organizations depend on real-time web data infrastructure, but 90% feel boxed in by various restrictions. Companies are increasingly developing technical approaches to navigate these constraints. Lenchner draws this metaphor: “Think of the trained model as intelligence and relevant data as knowledge. A powerful intelligence layer sitting on top of a hollow knowledge layer is like a genius who knows nothing—useless in practice. Intelligence and knowledge have to come together.” The promise of new infrastructure A new layer of web data infrastructure can address this developing need for stronger AI inputs by enabling discovery of data, real-time access, and tailoring to a specific context. As Lechner describes it, “It’s all about collecting data at scale, super-low latency, without being blocked.” Rather than relying on increased computing power, this type of platform emulates human browsing behavior to access available content and transform raw code into structured data feeds. It can work with websites that might not interact with traditional scraping tools, such as those heavy in JavaScript, or with aggressive antibot software. As Lenchner explains, “It’s basically having infrastructure that can mimic a web user with identifying information—IP address, location, and 1,000 more parameters. And at scale. Think of doing that 80 billion times a day for millions of websites. And every single time, you are looking exactly as the website expects you to look.” Of course, continuous retrieval introduces new data governance challenges. To address them, platforms can enforce strict compliance protocols aligned with global privacy frameworks, such as the EU’s General Data Protection Regulation (GDPR) and the California Consumer Privacy Act (CCPA). They can also be limited to openly accessible, public information, avoiding paywalls or private logins. Any networks used can be vetted and consent-based, and incentives can be provided to owners of IP addresses. In this way, systems can be designed to comply with tightening regulation. Such complex capabilities do not come easy. “When this is critical infrastructure for a company,” Lenchner says, “doing it in-house becomes a full-time engineering problem that competes with the actual AI work.” Addressing this complexity requires organizations to commit significant resources, leading many to seek specialized platforms designed specifically for data retrieval, orchestration, and observability. Infrastructure for the real world Real-time data retrieval is changing what AI
The emergence of the web data infrastructure layer for AI 投稿を読む »
In this article, you will learn why a large context window is not the same thing as agent memory, and how techniques like retrieval, compression,…
Context Windows Are Not Memory: What AI Agent Developers Need to Understand 投稿を読む »
We use cookies to improve your experience and performance on our website. You can learn more at プライバシーポリシー and manage your privacy settings by clicking Settings.
