
How robotic hives and AI are lowering the risk of bee colony collapse
The unit — dubbed a BeeHome — is an industrial upgrade from the standard wooden beehives, all clad in white metal and solar panels. Inside sits a high-tech scanner and robotic arm powered by artificial intelligence. Roughly 300,000 of these units are in use across the U.S., scattered across fields of almond, canola, pistachios and other crops that require pollination to grow.
It's not exactly the romantic vision of a beehive or beekeeper lodged in the cultural consciousness, but then that's not what matters; keeping bees alive does. And Beewise's units do that dramatically better than the standard hive, providing constant insights on colony health and the ability to provide treatment should it start to falter.
The U.S. has observed a startling uptick in the number of die-offs since the mid-2000s as beekeepers have struggled to keep pace with the rise of disease-carrying mites, climate extremes and other stressors that can wipe out colonies. That's endangering billions of dollars in crops from almonds to avocados that rely on the pollinators. This past year saw the worst colony losses on record. Beewise has raised nearly $170 million, including a $50 million Series D earlier this month, and it has a plan to change the industry.
AI and robotics are able to replace "90% of what a beekeeper would do in the field,' said Beewise Chief Executive Officer and co-founder Saar Safra. The question is whether beekeepers are willing to switch out what's been tried and true equipment.
Ultimately, the fate of humans is tied to that of bees. Roughly 75% of crops require pollinators, with nuts and fruits particularly dependent. While other species of bees and insects can play a role, they can't replace honeybees.
"There would essentially be no crop without the bees,' said Zac Ellis, the senior director of agronomy at OFI, a global food and ingredient seller.
Equipped with a camera and a robotic arm, BeeHomes are able to regularly snap images of the frames inside. |
Bloomberg
The beehive hasn't seen much technological innovation in 170 years. The Langstroth hive, named after the American reverend who patented it in 1852, is a simple wooden box with frames that can house the queen and her worker bees, larvae and honey.
"Langstroth hives are easy to work with, break down, build up, manipulate frames, make splits' and move, said Priya Chakrabarti Basu, a Washington State University bee researcher.
These boxes are the backbone of the agriculture industry and the high-value crops that are heavily reliant on the 2.5 million commercial hives that crisscross the U.S. on semi-trailers. Beekeepers with thousands of hives will travel from as far away as Florida to provide pollination services for California's $3.9 billion almond crop in spring before moving on to other states and crops.
"Almonds are one of the largest pollination events in the world,' said Ellis, who uses Beewise's hives on 30% of the acres he manages. "Typically, a grower needs two hives per acre,' each with up to 40,000 bees.
Pollinating the 10,000 acres of almonds, walnuts and pistachios he oversees requires millions of bees doing the brunt of the pollination work.
The number of hives and demand have created a problem, though: Beekeepers are only able to check on their colonies' health every week or two. But a growing number of threats to bees means entire colonies can be wiped out or weakened past the point of no return in just a few days.
Toxic pesticides, a changing climate and a sharp uptick in the invasive, disease-transmitting varroa mite since the 1980s have contributed to the rise of what's known as colony collapse disorder. The exact role each of these issues plays in wiping out colonies is unclear, but they are also likely interacting with each other to take a toll.
A hive contains up to 15 frames inside the BeeHome. |
Bloomberg
"You are rarely going to find a bee who is only, for example, stressed by a mite or a bee who's stressed by a disease only or a bee who's only stressed by poor nutrition,' Chakrabarti Basu said. "It is always a combination.'
The impacts, though, are clear. From the 12-month period starting last April, more than 56% of commercial colonies were wiped out, according to the Apiary Inspectors of America. Beekeepers have taken a major economic hit as a result: Between last June and March, colony losses cost beekeepers an estimated $600 million, according to the Honey Bee Health Coalition.
While a new hive design alone isn't enough to save bees, Beewise's robotic hives help cut down on losses by providing a near-constant stream of information on colony health in real time — and give beekeepers the ability to respond to issues. Equipped with a camera and a robotic arm, they're able to regularly snap images of the frames inside the BeeHome, which Safra likened to an MRI. The amount of data they capture is staggering.
Each frame contains up to 6,000 cells where bees can, among other things, gestate larvae or store honey and pollen. A hive contains up to 15 frames and a BeeHome can hold up to 10 hives, providing thousands of datapoints for Beewise's AI to analyze.
While a trained beekeeper can quickly look at a frame and assess its health, AI can do it even faster, as well as take in information on individual bees in the photos. Should AI spot a warning sign, such as a dearth of new larvae or the presence of mites, beekeepers will get an update on an app that a colony requires attention. The company's technology earned it a BloombergNEF Pioneers award earlier this year.
Beewise co-founder and CEO Saar Safra |
Bloomberg
"There's other technologies that we've tried that can give us some of those metrics as well, but it's really a look in the rearview mirror,' Ellis said. "What really attracted us to Beewise is their ability to not only understand what's happening in that hive, but to actually act on those different metrics.'
That includes administering medicine and food as well as opening and closing vents to regulate temperature or protect against pesticide spraying. Safra noted that after two hurricanes hit Florida last year, BeeHomes in the state were still operational while many wooden hives were destroyed.
That durability and responsiveness has Ellis convinced on expanding their use. Today, BeeHomes are on 30% of his acres, but he said within three years, they're aiming for 100% coverage. Whether other growers and beekeepers are as keen to make the switch remains to be seen, though, given nearly two centuries of loyalty to the Langstroth design.
The startup wants to more than triple the number of BeeHomes in use, reaching 1 million in three years.
"We're in a race against time,' Safra said. "We might have the best product on planet earth in 15 years, but it doesn't matter' if there aren't any bees left.
Ellis likened the hives to a Ritz-Carlton for pollinators. The five-star stay appears to suit bees well: Beewise says its units — which it leases to provide pollination services at what it says are market rates — have seen colony losses of around 8%. That's a major drop compared to the average annual loss rate of more than 40%, according to Apiary Inspectors of America, a group that tracks colony health.
The Beewise hive exterior is made of white metal and solar panels. |
Bloomberg
"The asset is the bees, that's the revenue-generating asset,' said Safra, noting that losing more than 40% of those assets makes it hard for businesses to cover labor to maintain hives, trailers to transport them and other fixed costs.
Beewise expects to have $100 million in revenue this year, and Safra said it's a year away from profitability. The company declined to share the valuation for its Series D.
It has competition in the bee-saving technology realm. Some companies like Dalan Animal Health are developing vaccines to protect bees against disease. BeeHero and Beeflow (sensing a pattern?) are among those that provide sensors for monitoring health in hives and fields. Both can help improve outcomes at Langstroth hives, but they still require regular beekeeper maintenance.
Chakrabarti Basu from Washington State and her colleagues are also working on using AI to detect bees entering hives. "The more data sets we can give, the better it'll be trained,' she said. "Pattern recognition — it could be monitoring a brood frame, it could be looking at anything for estimating colony health or any aspect of colony health — I think AI will probably get better at it.'

Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles


Yomiuri Shimbun
3 hours ago
- Yomiuri Shimbun
Google's AI Overviews Hit by EU Antitrust Complaint from Independent Publishers
BRUSSELS, July 4 (Reuters) – Alphabet's GOOGL.O Google has been hit by an EU antitrust complaint over its AI Overviews from a group of independent publishers, which has also asked for an interim measure to prevent allegedly irreparable harm to them, according to a document seen by Reuters. Google's AI Overviews are AI-generated summaries that appear above traditional hyperlinks to relevant webpages and are shown to users in more than 100 countries. It began adding advertisements to AI Overviews last May. The company is making its biggest bet by integrating AI into search but the move has sparked concerns from some content providers such as publishers. The Independent Publishers Alliance document, dated June 30, sets out a complaint to the European Commission and alleges that Google abuses its market power in online search. 'Google's core search engine service is misusing web content for Google's AI Overviews in Google Search, which have caused, and continue to cause, significant harm to publishers, including news publishers in the form of traffic, readership and revenue loss,' the document said. It said Google positions its AI Overviews at the top of its general search engine results page to display its own summaries which are generated using publisher material and it alleges that Google's positioning disadvantages publishers' original content. 'Publishers using Google Search do not have the option to opt out from their material being ingested for Google's AI large language model training and/or from being crawled for summaries, without losing their ability to appear in Google's general search results page,' the complaint said. The Commission declined to comment. The UK's Competition and Markets Authority confirmed receipt of the complaint. Google said it sends billions of clicks to websites each day. 'New AI experiences in Search enable people to ask even more questions, which creates new opportunities for content and businesses to be discovered,' a Google spokesperson said. The Independent Publishers Alliance's website says it is a nonprofit community advocating for independent publishers, which it does not name. The Movement for an Open Web, whose members include digital advertisers and publishers, and British non-profit Foxglove Legal Community Interest Company, which says it advocates for fairness in the tech world, are also signatories to the complaint. They said an interim measure was necessary to prevent serious irreparable harm to competition and to ensure access to news. Google said numerous claims about traffic from search are often based on highly incomplete and skewed data. 'The reality is that sites can gain and lose traffic for a variety of reasons, including seasonal demand, interests of users, and regular algorithmic updates to Search,' the Google spokesperson said. Foxglove co-executive director Rosa Curling said journalists and publishers face a dire situation. 'Independent news faces an existential threat: Google's AI Overviews,' she told Reuters. 'That's why with this complaint, Foxglove and our partners are urging the European Commission, along with other regulators around the world, to take a stand and allow independent journalism to opt out,' Curling said. The three groups have filed a similar complaint and a request for an interim measure to the UK competition authority. The complaints echoed a U.S. lawsuit by a U.S. edtech company which said Google's AI Overviews is eroding demand for original content and undermining publishers' ability to compete that have resulted in a drop in visitors and subscribers.


Japan Today
5 hours ago
- Japan Today
Tech tip: Get the most out of ChatGPT and other AI chatbots with better prompts
By KELVIN CHAN If you're using ChatGPT but getting mediocre results, don't blame the chatbot. Instead, try sharpening up your prompts. Generative AI chatbots such as OpenAI's ChatGPT, Google's Gemini and Anthropic's Claude have become hugely popular and embedded into daily life for many users. They're powerful tools that can help us with so many different tasks. What you shouldn't overlook, however, is that a chatbot's output depends on what you tell it to do, and how. There's a lot you can do to improve the prompt — also known as the request or query — that you type in. Here are some tips for general users on how to get higher quality chatbot replies, based on tips from the AI model makers: ChatGPT can't read your mind. You need to give it clear and explicit instructions on what you need it to do. Unlike a standard Google search, you can't just ask for an answer based on some keywords. And you'll need to do more than just tell it to, say, 'design a logo' because you'll end up with a generic design. Flesh it out with details on the company that the logo is for, the industry it will be used in and the design style you're going for. 'Ensure your prompts are clear, specific, and provide enough context for the model to understand what you are asking,' ChatGPT maker OpenAI advises on its help page. 'Avoid ambiguity and be as precise as possible to get accurate and relevant responses.' Think of using a chatbot like holding a conversation with a friend. You probably wouldn't end your chat after the first answer. Ask follow-up questions or refine your original prompt. OpenAI's advice: 'Adjust the wording, add more context, or simplify the request as needed to improve the results.' You might have to have an extended back-and-forth that elicits better output. Google advises that you'll need to try a 'few different approaches' if you don't get what you're looking for the first time. 'Fine-tune your prompts if the results don't meet your expectations or if you believe there's room for improvement,' Google recommends in its prompting guide for Gemini. 'Use follow-up prompts and an iterative process of review and refinement to yield better results.' When making your request, you can also ask an AI large language model to respond in a specific voice or style. 'Words like formal, informal, friendly, professional, humorous, or serious can help guide the model,' OpenAI writes. You also tell the chatbot the type of person the response is aimed at. These parameters will help determine the chatbot's overall approach to its answer, as well as the tone, vocabulary and level of detail. For example, you could ask ChatGPT to describe quantum physics in the style of a distinguished professor talking to a class of graduate students. Or you could ask it to explain the same topic in the voice of a teacher talking to a group of schoolchildren. However, there's plenty of debate among AI experts about these methods. On one hand, they can make answers more precise and less generic. But an output that adopts an overly empathetic or authoritative tone raises concerns about the text sounding too manipulative. Give the chatbot all the background behind the reason for your request. Don't just ask: 'Help me plan a weeklong trip to London.' ChatGPT will respond with a generic list of London's greatest hits: historic sites on one day, museums and famous parks on another, trendy neighborhoods and optional excursions to Windsor Castle. It's nothing you couldn't get from a guidebook or travel website, but just a little better organized. But if, say, you're a theatre-loving family, try this: 'Help me plan a weeklong trip to London in July, for a family of four. We don't want too many historic sites, but want to see a lot of West End theatre shows. We don't drink alcohol so we can skip pubs. Can you recommend mid-range budget hotels where we can stay and cheap places to eat for dinner?" This prompt returns a more tailored and detailed answer: a list of four possible hotels within walking distance of the theater district, a seven-day itinerary with cheap or low-cost ideas for things to do during the day, suggested shows each evening, and places for an affordable family dinner. You can tell any of the chatbots just how extensive you want the answer to be. Sometimes, less is more. Try nudging the model to provide clear and succinct responses by imposing a limit. For example, tell the chatbot to reply with only 300 words, or to come up with five bullet points. Want to know all that there is to know about quantum physics? ChatGPT will provide a high-level 'grand tour' of the topic that includes terms like wavefunctions and qubits. But ask for a 150-word explanation and you'll get an easily digestible summary about how it's the science of the tiniest particles that also underpins a lot of modern technology like lasers and smartphones. © Copyright 2025 The Associated Press. All rights reserved. This material may not be published, broadcast, rewritten or redistributed without permission.

Japan Times
a day ago
- Japan Times
A ‘Sputnik' moment in the global AI race
When Chinese AI startup DeepSeek unveiled the open-source large language model DeepSeek-R1 in January, many referred to it as the "AI Sputnik shock" — a reference to the monumental significance of the Soviet Union's 1957 launch of the first satellite into orbit. Much remains uncertain about DeepSeek's LLM and its capabilities should not be overestimated — but its release nevertheless has sparked intense discussion about its superiority especially in terms of cost. DeepSeek claims that its model possesses reasoning abilities on par with or even superior to OpenAI's leading models, with training costs at less than one-tenth of OpenAI's — reportedly just $5.6 million — largely due to the use of NVIDIA's lower-cost H800 GPUs rather than the more powerful H200 or H100 models. Tech giants like Meta and Google have spent billions of dollars on high-performance GPUs to develop cutting-edge AI models. However, DeepSeek's ability to produce a high-performance AI model at a significantly lower cost challenges the prevailing belief that computational power—determined by the number and quality of GPUs—is the primary driver of AI performance.